This paper explores a spectral vector-based methodology on Landsat 8 bands of the visible wavelen... more This paper explores a spectral vector-based methodology on Landsat 8 bands of the visible wavelengths, that is deep-blue (1) to shortwave infrared (7), to improve the urban land features classification. Using two different ratio models, based on two and three bands' combinations in the cloud environment of Google Earth Engine, the Uncertainty reducing Spectral Vector (USV<sub>r</sub>), the Onward Continuous Spectral Vector (OSV<sub>c</sub>) and the Onward Discontinuous Spectral Vector (OSV<sub>d</sub>) are proposed as new entries for the land use land cover (LULC) classification. Two different sizes of arrays are built, i.e. 42 vectors and 15 vectors corresponding to the same number of derivative bands and new pixels′ values. A decision tree is built in <i>J.48</i> and applied to select the most suitable derivative bands for the analysis. Hereafter, the selected ones are stacked and submitted to five machine learning classifiers using a supervised process, namely, Classification and Regression Trees (CART), Random Forest (RF) Gradient Boosting (GBR), Support Vector Machine (SVM) and Minimum Distance (MD). This method was tested in the two cities of Bamenda and Foumban in west-Cameroon highlands, due to their good representativeness of tropical hilly urban areas' spatial heterogeneity. The results are satisfying for 4/5 classifiers, up to 87% Overall Accuracy, OA, for 0.82 kappa coefficient, KC, in Bamenda, while combining SVM/OSV<sub>d</sub>. Whereas, in Foumban, the classifiers perform up to 85%OA and 0.78 KC for the combination SVM/USV<sub>r</sub>. Only the MD classifier has always performed below 80%OA. The process has been found better than performing classifiers directly on the multispectral (MS) image, by providing more possibilities of hidden spectral indices not yet explored, as far as we know, to detect and discriminate between LULC features, plus an accurate extraction of human settlements.
Many cities of developing countries sprawl out because there is barely a planning process, genera... more Many cities of developing countries sprawl out because there is barely a planning process, generally with a mixed housing system. Moreover, most local studies mainly use free satellite images of medium to low/coarse resolution, with low accuracy of results. The main goal of this poster was to implement a Geospatial method to assess the metabolism (transformation) of a mid-urban/mid-rural city. Landsat images of 1987, 2003 and 2019 was used. The first point has been to enhance the built-up detection by proposing the Modified New Built-up Index (MNBI), as the root squared of the sum of squares of the New Built-up Index (NBI; Jieli C. et al., 2010) and the Landsat shortwave infrared two (SWIR2) band. Further, the Land Surface Temperature (LST) was computed and a linear regression was made with the MNBI, Bare soil index (Rikimaru A. et al, 2002) and the Normalized Difference Vegetation Index (Rouse, J.W. et al, 1974). Rate of change (RC) and change average (CA) of slopes and intercepts from the linear equation of LST regressed by BI and NDVI , have helped to better understand the urban metabolism. Finally, RC and CA values have been used in a last equation applied to the 2019 LST, to identify and classify areas of mitigation, i.e. replanting.
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2020
This paper addresses the remote sensing challenging field of urban mixed pixels on a medium spati... more This paper addresses the remote sensing challenging field of urban mixed pixels on a medium spatial resolution satellite data. The tentatively named Normalized Difference Built-up and Surroundings Unmixing Index (NDBSUI) is proposed by using Landsat-8 Operational Land Imager (OLI) bands. It uses the Shortwave Infrared 2 (SWIR2) as the main wavelength, the SWIR1 with the red wavelengths, for the built-up extraction. A ratio is computed based on the normalization process and the application is made on six cities with different urban and environmental characteristics. The built-up of the experimental site of Yaoundé is extracted with an overall accuracy of 95.51% and a kappa coefficient of 0.90. The NDBSUI is validated over five other sites, chosen according to Cameroon's bioclimatic zoning. The results are satisfactory for the cities of Yokadouma and Kumba in the bimodal and monomodal rainfall zones, where overall accuracies are up to 98.9% and 97.5%, with kappa coefficients of 0.88 and 0.94 respectively, although these values are close to those of three other indices. However, in the cities of Foumban, Ngaoundéré and Garoua, representing the western highlands, the high Guinea savannah and the Sudano-sahelian zones where built-up is more confused with soil features, overall accuracies of 97.06%, 95.29% and 74.86%, corresponding to 0.918, 0.89 and 0.42 kappa coefficients were recorded. Difference of accuracy with EBBI, NDBI and UI are up to 31.66%, confirming the NDBSUI efficiency to automate built-up extraction and unmixing from surrounding noises with less biases.
In this paper, Analytic Hierarchy Process (AHP) is used as the method of criteria choice and clas... more In this paper, Analytic Hierarchy Process (AHP) is used as the method of criteria choice and classification for the delimitation of a protected area. Starting from the hypothesis that the space define for protected areas of Cameroon must integrate in priority wildlife, the density of spatial distribution of medium and large mammals in the Campo Ma'an National Park is used here as the main criterion analysis. Its combination in raster mode with other layers of information such as vegetation, streams, tracks, surrounding localities and areas of human activities provides a map on which the park area increased by 88.6%. This result is discussed on the basis of socio-spatial impact which is assessed in terms of three levels of relocation of the surrounding villages and their populations.
International Journal of Advanced Remote Sensing and GIS, 2018
The aim of this study is to assess the land use/land cover (LULC) inter-seasonal changes along th... more The aim of this study is to assess the land use/land cover (LULC) inter-seasonal changes along the Cameroonian shores of Lake Chad and its hinterland using the four generations of Landsat sensors images of MSS, TM, ETM+ and OLI. Identification of land use/land cover inter-seasonal changes is based on classification by Support Vector Machines (SVMs) algorithm. Three major spatial classes of objects are identified: open water, vegetation and marshlands, and bare soils. The results show that, bare soils class has the higher rate, and can reach 67.57% of the study area extent. Moreover, land use/land cover change from one season to another or from one decade to another can be closely linked to evolution of climate conditions. Then open water areas vary little with rate of 1.94% and 7.6% for inter-seasonal changes, and rate of 5.62% and 63.05% for inter-annual changes. Compared to open water, vegetation and marshlands has the most important variation, that is 583.59%. In addition, proportions of bare soils that vary are different between dry seasons and rainy seasons, with a lower and a higher rate of 5.38% and 82.8%. This leads to the conclusion that occupation in the study area is dominated by bare soils principally; occupation class that is most affected by changes is vegetation and marshlands followed by bare soils and then open water and marshlands.
ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2020
Drought affects all human activities and ecosystems. Nearly 40 percent of the world's population ... more Drought affects all human activities and ecosystems. Nearly 40 percent of the world's population inhabit Drylands, and they depend on agriculture for their food, security and livelihoods. Among the remote sensing indices developed, the Land Surface General Drought Index (LSGDI) was recently proposed. This paper proposes an improved model of LSGDI to face the issue of drought in semi-arid and arid regions. The experiment was conducted for the Maga's floodplain, in North-Cameroon. The method uses satellite images of Landsat in 1987, 2003 and 2018, for January and March or April, corresponding to the middle and the end of the dry season. A Vegetation Moisture Index (VMI) and a Normalized Difference Soil Drought Index (NDSoDI) are both developed. On an orthogonal plan, their projections give a drought line that expresses the improved LSGDI (LSGDI2) as the root sum square of the NDSoDI and the VMI. The LSGDI2 results are ranged in [0.09-0.14] interval, which is used to define the threshold and ease the qualifiers for drought classes. The visual patterns easily match the sandy areas of the original Landsat images with the highest values, while the vegetation and water areas match the lowest values. Compared with the LSGDI and Second Modified Perpendicular drought Index (MPDI1), the new index reflectance values are higher. Finally, although LSGDI2 curve's evolution follows the NDSoDI one at 94%, the new spectral index values depends on the both components, helping to map highest values of drought and moisture in Maga's floodplain, for a sustainable rice culture expansion.
International Journal of Advanced Remote Sensing and GIS, 2016
This paper aimed to assess the status of land degradation in arid and semi-arid areas based on a ... more This paper aimed to assess the status of land degradation in arid and semi-arid areas based on a correlation analysis between spectral and statistical neo-bands. The methodology uses vegetation and soil spectral indices as the second Modified Soil Adjusted Vegetation Index (MSAVI2), Normalized Difference Bare Soil Index (NDBSI), Texture Index (NDTeI), Crust Index (CI), Top Soil Grain Size Index (GSI), Normalized Difference Sand Dune Index (NDSDI) and the first Specific Principal Component of the red, near infrared, shortwave infrared bands stacking (SPC1 R-NIR-SWIR1-SWIR2). The vegetation is considered here as the main object of soil sub-surface. Thus after all the spectral and the statistic neo-bands are performed on Landsat8 OLI sensor image, a linear regression is generated to assess their correlation with MSAVI2. Based on the visual interpretation and the regression curves the results show that the determination coefficient R 2 and the P values all significant as less than 0.0001. Each neo-band is weighted with its R 2 to improve its contribution to the model and the synthesis image obtained enhances the land degradation sensing in six classes; these are respectively named as ''severe'' (3139 km 2), ''high'' (6763 km 2), ''moderate'' (8341 km 2), ''low'' (7454 km 2), ''very low'' (6947 km 2) and ''close to nil'' (5437 km 2). This last image is summed with population layer to produce a decision map helpful for further government decision. At the end the degradation image has given interesting results for the detection of land degradation comparatively to derivation and comparison of individual indices.
This chapter proposes a remote sensing multi-angles methodology to assess the transition at the i... more This chapter proposes a remote sensing multi-angles methodology to assess the transition at the interface of the forest-savanna land cover. On Sentinel2-A median images of successive dry seasons, three referential and nine analytical spectral indices were computed. The change vector analysis (CVA) was performed, selecting further one magnitude per index. The averaged moving standard deviation index (aMSDI) was proposed to compare spatial intensity of anomalies among selected CVA, and then statistically assessed through spatial and no-spatial autoregression tests. The cross-correlation and simple linear combination (SCL) computations spotted the overall anomaly extent. Three machine learning algorithms, i.e., classification and regression trees (CART), random forest (RF), and support vector machine (SVM), helped mapping the distribution of each specie. As result, the CVA confirmed each index ability to add new information. The aMSDI gave the harmonized interval [0–0.083] among CVA, c...
This paper explores a spectral vector-based methodology on Landsat 8 bands of the visible wavelen... more This paper explores a spectral vector-based methodology on Landsat 8 bands of the visible wavelengths, that is deep-blue (1) to shortwave infrared (7), to improve the urban land features classification. Using two different ratio models, based on two and three bands' combinations in the cloud environment of Google Earth Engine, the Uncertainty reducing Spectral Vector (USV<sub>r</sub>), the Onward Continuous Spectral Vector (OSV<sub>c</sub>) and the Onward Discontinuous Spectral Vector (OSV<sub>d</sub>) are proposed as new entries for the land use land cover (LULC) classification. Two different sizes of arrays are built, i.e. 42 vectors and 15 vectors corresponding to the same number of derivative bands and new pixels′ values. A decision tree is built in <i>J.48</i> and applied to select the most suitable derivative bands for the analysis. Hereafter, the selected ones are stacked and submitted to five machine learning classifiers using a supervised process, namely, Classification and Regression Trees (CART), Random Forest (RF) Gradient Boosting (GBR), Support Vector Machine (SVM) and Minimum Distance (MD). This method was tested in the two cities of Bamenda and Foumban in west-Cameroon highlands, due to their good representativeness of tropical hilly urban areas' spatial heterogeneity. The results are satisfying for 4/5 classifiers, up to 87% Overall Accuracy, OA, for 0.82 kappa coefficient, KC, in Bamenda, while combining SVM/OSV<sub>d</sub>. Whereas, in Foumban, the classifiers perform up to 85%OA and 0.78 KC for the combination SVM/USV<sub>r</sub>. Only the MD classifier has always performed below 80%OA. The process has been found better than performing classifiers directly on the multispectral (MS) image, by providing more possibilities of hidden spectral indices not yet explored, as far as we know, to detect and discriminate between LULC features, plus an accurate extraction of human settlements.
International Journal of Advanced Remote Sensing and GIS, 2016
This paper aimed to assess the status of land degradation in arid and semi-arid areas based on a ... more This paper aimed to assess the status of land degradation in arid and semi-arid areas based on a correlation analysis between spectral and statistical neo-bands. The methodology uses vegetation and soil spectral indices as the second Modified Soil Adjusted Vegetation Index (MSAVI2), Normalized Difference Bare Soil Index (NDBSI), Texture Index (NDTeI), Crust Index (CI), Top Soil Grain Size Index (GSI), Normalized Difference Sand Dune Index (NDSDI) and the first Specific Principal Component of the red, near infrared, shortwave infrared bands stacking (SPC1R-NIR-SWIR1-SWIR2). The vegetation is considered here as the main object of soil sub-surface. Thus after all the spectral and the statistic neo-bands are performed on Landsat8 OLI sensor image, a linear regression is generated to assess their correlation with MSAVI2. Based on the visual interpretation and the regression curves the results show that the determination coefficient R2 and the P values all significantas less than 0.0001. ...
Oil Palm (Elaeis guineensis Jacq.) has recorded a boom production the last decades and its main p... more Oil Palm (Elaeis guineensis Jacq.) has recorded a boom production the last decades and its main productive zone is inside the tropics that meet the best biophysical conditions. Investors as well as geospatial practitioners are increasingly interested on the best growing and harvesting conditions. So said, the aim of this paper is to select the best oil palm planting site through the best methods combination. The study area is the district of Njimom located in the west-Cameroon, transitional between the equatorial and the climatic zones. In the same GIS environment, the Weighted Linear Combination (WLC) and Fuzzy Analytic Hierarchy Process (FAHP) respectively highlight the subtle differences between capability and suitability, while the Utility Function (UF) helps to assess the consideration of sustainability aspects. The first results consist in eight layers representing natural conditions, that is rainfall, temperatures, sunshine, slope, elevation, soil richness, soil moisture and forest cover, recoded in six classes ranked from 5 to 0 according to the FAO standardised scale. They are crossed using the straightforward method of WLC to give the "Capability layer". The second results consist in three layers related to the social-economical constraints for production, as built-up How to cite this paper: Mfondoum, A.H.N.,
This paper addresses the remote sensing challenging field of urban mixed pixels on a medium spati... more This paper addresses the remote sensing challenging field of urban mixed pixels on a medium spatial resolution satellite data. The tentatively named Normalized Difference Built-up and Surroundings Unmixing Index (NDBSUI) is proposed by using Landsat-8 Operational Land Imager (OLI) bands. It uses the Shortwave Infrared 2 (SWIR2) as the main wavelength, the SWIR1 with the red wavelengths, for the built-up extraction. A ratio is computed based on the normalization process and the application is made on six cities with different urban and environmental characteristics. The built-up of the experimental site of Yaoundé is extracted with an overall accuracy of 95.51% and a kappa coefficient of 0.90. The NDBSUI is validated over five other sites, chosen according to Cameroon's bioclimatic zoning. The results are satisfactory for the cities of Yokadouma and Kumba in the bimodal and monomodal rainfall zones, where overall accuracies are up to 98.9% and 97.5%, with kappa coefficients of 0.88 and 0.94 respectively, although these values are close to those of three other indices. However, in the cities of Foumban, Ngaoundéré and Garoua, representing the western highlands, the high Guinea savannah and the Sudano-sahelian zones where built-up is more confused with soil features, overall accuracies of 97.06%, 95.29% and 74.86%, corresponding to 0.918, 0.89 and 0.42 kappa coefficients were recorded. Difference of accuracy with EBBI, NDBI and UI are up to 31.66%, confirming the NDBSUI efficiency to automate built-up extraction and unmixing from surrounding noises with less biases.
Drought affects all human activities and ecosystems. Nearly 40 percent of the world's population ... more Drought affects all human activities and ecosystems. Nearly 40 percent of the world's population inhabit Drylands, and they depend on agriculture for their food, security and livelihoods. Among the remote sensing indices developed, the Land Surface General Drought Index (LSGDI) was recently proposed. This paper proposes an improved model of LSGDI to face the issue of drought in semi-arid and arid regions. The experiment was conducted for the Maga's floodplain, in North-Cameroon. The method uses satellite images of Landsat, corresponding to the middle and the end of the dry season. A Vegetation Moisture Index (VMI) and a Normalized Difference Soil Drought Index (NDSoDI) are both developed. On an orthogonal plan, their projections give a drought line that expresses the improved LSGDI (LSGDI2) as the root sum square of the NDSoDI and the VMI. The LSGDI2 results are ranged in [0.09-0.14] interval, which is used to define the threshold and ease the qualifiers for drought classes. The visual patterns easily match the sandy areas of the original Landsat images with the highest values, while the vegetation and water areas match the lowest values. Compared with the LSGDI and Second Modified Perpendicular drought Index (MPDI1), the new index reflectance values are higher. Finally, although LSGDI2 curve's evolution follows the NDSoDI one at 94%, the new spectral index values depends on the both components, helping to map highest values of drought and moisture in Maga's floodplain, for a sustainable rice culture expansion.
International Journal of Advanced Remote Sensing and GIS, 2018
The aim of this study is to assess the land use/land cover (LULC) inter-seasonal changes along th... more The aim of this study is to assess the land use/land cover (LULC) inter-seasonal changes along the Cameroonian shores of Lake Chad and its hinterland using the four generations of Landsat sensors images of MSS, TM, ETM+ and OLI. Identification of land use/land cover inter-seasonal changes is based on classification by Support Vector Machines (SVMs) algorithm. Three major spatial classes of objects are identified: open water, vegetation and marshlands, and bare soils. The results show that, bare soils class has the higher rate, and can reach 67.57% of the study area extent. Moreover, land use/land cover change from one season to another or from one decade to another can be closely linked to evolution of climate conditions. Then open water areas vary little with rate of 1.94% and 7.6% for inter-seasonal changes, and rate of 5.62% and 63.05% for inter-annual changes. Compared to open water, vegetation and marshlands has the most important variation, that is 583.59%. In addition, proportions of bare soils that vary are different between dry seasons and rainy seasons, with a lower and a higher rate of 5.38% and 82.8%. This leads to the conclusion that occupation in the study area is dominated by bare soils principally; occupation class that is most affected by changes is vegetation and marshlands followed by bare soils and then open water and marshlands.
Oil Palm (Elaeis guineensis Jacq.) has recorded a boom production the last decades and its main p... more Oil Palm (Elaeis guineensis Jacq.) has recorded a boom production the last decades and its main productive zone is inside the tropics that meet the best biophysical conditions. Investors as well as geospatial practitioners are increasingly interested on the best growing and harvesting conditions. So said, the aim of this paper is to select the best oil palm planting site through the best methods combination. The study area is the district of Njimom located in the west-Cameroon, transitional between the equatorial and the climatic zones. In the same GIS environment, the Weighted Linear Combination (WLC) and Fuzzy Analytic Hierarchy Process (FAHP) respectively highlight the subtle differences between capability and suitability, while the Utility Function (UF) helps to assess the consideration of sustainability aspects. The first results consist in eight layers representing natural conditions, that is rainfall, temperatures, sunshine, slope, elevation, soil richness, soil moisture and forest cover, recoded in six classes ranked from 5 to 0 according to the FAO standardised scale. They are crossed using the straightforward method of WLC to give the "Capability layer". The second results consist in three layers related to the social-economical constraints for production, as built-up area, distance to road and distance to rivers. These layers are recoded in binary with 1 and 0, they are weighted by the FAHP priority vector and membership approach to give the “suitability layer”. Then, the number of occurrences of each aspect of the sustainability is counted in each of the two preceding processes to perform the UF. The resulting value, that is 0.542 for the capability process and 0.315 for the suitability process, serves to weight their respective layers, and their sum gives the final map with the best oil palm site planting in the northern part of the study area, on about 34,950 ha, representing 44.8% of Njimom district.
IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES, 2019
This paper proposes a tool resulting from the merging of the Eisenhower matrix and Analytical Hie... more This paper proposes a tool resulting from the merging of the Eisenhower matrix and Analytical Hierarchy Process to strengthen the prioritization of actions for managers. The case study is from a Ph.D. thesis work, which the first phase is to produce the candidate's schedule. An Eisenhower matrix is designed and priorities are affected to corresponding quadrant, based on the Priority Quotient (PQ) results. As it is too expert, the second phase, concerning a chapter about the calculation of an Accessibility Index is about designing easier readable tool directly usable by the managers to undertake their actions, saving time and investment. A 4*4 matrix called Actions of Governance Matrix (AGM) accompanied by a Partial Priority Quotient (PPQ) scale issued from the AHP process is the resulting tool.
IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES, 2019
The aim of this paper is to assess the impact of Participatory GIS (PGIS) in the process of decis... more The aim of this paper is to assess the impact of Participatory GIS (PGIS) in the process of decision-making in a developing country context. This practice is an alternative process between GIS and Participatory Learning and Action (PLA) methods. It highlights specificities in agriculture and livestocks, health and tourism whose different departments are adjusting to government's efforts to solve local issues while including local populations. The main methods used here are cognitive mapping and focus groups with GIS facilitators. The results obtained are Counting Zones, Health Areas and Touristic Spatial Units delimitations confined or not inside the different district's boundaries, but consensually validated by all the stakeholders. It confirms that the Indigenous System of knowledge (ISK), combined to GIS expertise and administrative coordination, is dynamic in supporting the Cameroon's advances on the field of proximity governance and territorial planning.
This paper explores a spectral vector-based methodology on Landsat 8 bands of the visible wavelen... more This paper explores a spectral vector-based methodology on Landsat 8 bands of the visible wavelengths, that is deep-blue (1) to shortwave infrared (7), to improve the urban land features classification. Using two different ratio models, based on two and three bands&amp;#39; combinations in the cloud environment of Google Earth Engine, the Uncertainty reducing Spectral Vector (USV&amp;lt;sub&amp;gt;r&amp;lt;/sub&amp;gt;), the Onward Continuous Spectral Vector (OSV&amp;lt;sub&amp;gt;c&amp;lt;/sub&amp;gt;) and the Onward Discontinuous Spectral Vector (OSV&amp;lt;sub&amp;gt;d&amp;lt;/sub&amp;gt;) are proposed as new entries for the land use land cover (LULC) classification. Two different sizes of arrays are built, i.e. 42 vectors and 15 vectors corresponding to the same number of derivative bands and new pixels′ values. A decision tree is built in &amp;lt;i&amp;gt;J.48&amp;lt;/i&amp;gt; and applied to select the most suitable derivative bands for the analysis. Hereafter, the selected ones are stacked and submitted to five machine learning classifiers using a supervised process, namely, Classification and Regression Trees (CART), Random Forest (RF) Gradient Boosting (GBR), Support Vector Machine (SVM) and Minimum Distance (MD). This method was tested in the two cities of Bamenda and Foumban in west-Cameroon highlands, due to their good representativeness of tropical hilly urban areas&amp;#39; spatial heterogeneity. The results are satisfying for 4/5 classifiers, up to 87% Overall Accuracy, OA, for 0.82 kappa coefficient, KC, in Bamenda, while combining SVM/OSV&amp;lt;sub&amp;gt;d&amp;lt;/sub&amp;gt;. Whereas, in Foumban, the classifiers perform up to 85%OA and 0.78 KC for the combination SVM/USV&amp;lt;sub&amp;gt;r&amp;lt;/sub&amp;gt;. Only the MD classifier has always performed below 80%OA. The process has been found better than performing classifiers directly on the multispectral (MS) image, by providing more possibilities of hidden spectral indices not yet explored, as far as we know, to detect and discriminate between LULC features, plus an accurate extraction of human settlements.
Many cities of developing countries sprawl out because there is barely a planning process, genera... more Many cities of developing countries sprawl out because there is barely a planning process, generally with a mixed housing system. Moreover, most local studies mainly use free satellite images of medium to low/coarse resolution, with low accuracy of results. The main goal of this poster was to implement a Geospatial method to assess the metabolism (transformation) of a mid-urban/mid-rural city. Landsat images of 1987, 2003 and 2019 was used. The first point has been to enhance the built-up detection by proposing the Modified New Built-up Index (MNBI), as the root squared of the sum of squares of the New Built-up Index (NBI; Jieli C. et al., 2010) and the Landsat shortwave infrared two (SWIR2) band. Further, the Land Surface Temperature (LST) was computed and a linear regression was made with the MNBI, Bare soil index (Rikimaru A. et al, 2002) and the Normalized Difference Vegetation Index (Rouse, J.W. et al, 1974). Rate of change (RC) and change average (CA) of slopes and intercepts from the linear equation of LST regressed by BI and NDVI , have helped to better understand the urban metabolism. Finally, RC and CA values have been used in a last equation applied to the 2019 LST, to identify and classify areas of mitigation, i.e. replanting.
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2020
This paper addresses the remote sensing challenging field of urban mixed pixels on a medium spati... more This paper addresses the remote sensing challenging field of urban mixed pixels on a medium spatial resolution satellite data. The tentatively named Normalized Difference Built-up and Surroundings Unmixing Index (NDBSUI) is proposed by using Landsat-8 Operational Land Imager (OLI) bands. It uses the Shortwave Infrared 2 (SWIR2) as the main wavelength, the SWIR1 with the red wavelengths, for the built-up extraction. A ratio is computed based on the normalization process and the application is made on six cities with different urban and environmental characteristics. The built-up of the experimental site of Yaoundé is extracted with an overall accuracy of 95.51% and a kappa coefficient of 0.90. The NDBSUI is validated over five other sites, chosen according to Cameroon's bioclimatic zoning. The results are satisfactory for the cities of Yokadouma and Kumba in the bimodal and monomodal rainfall zones, where overall accuracies are up to 98.9% and 97.5%, with kappa coefficients of 0.88 and 0.94 respectively, although these values are close to those of three other indices. However, in the cities of Foumban, Ngaoundéré and Garoua, representing the western highlands, the high Guinea savannah and the Sudano-sahelian zones where built-up is more confused with soil features, overall accuracies of 97.06%, 95.29% and 74.86%, corresponding to 0.918, 0.89 and 0.42 kappa coefficients were recorded. Difference of accuracy with EBBI, NDBI and UI are up to 31.66%, confirming the NDBSUI efficiency to automate built-up extraction and unmixing from surrounding noises with less biases.
In this paper, Analytic Hierarchy Process (AHP) is used as the method of criteria choice and clas... more In this paper, Analytic Hierarchy Process (AHP) is used as the method of criteria choice and classification for the delimitation of a protected area. Starting from the hypothesis that the space define for protected areas of Cameroon must integrate in priority wildlife, the density of spatial distribution of medium and large mammals in the Campo Ma'an National Park is used here as the main criterion analysis. Its combination in raster mode with other layers of information such as vegetation, streams, tracks, surrounding localities and areas of human activities provides a map on which the park area increased by 88.6%. This result is discussed on the basis of socio-spatial impact which is assessed in terms of three levels of relocation of the surrounding villages and their populations.
International Journal of Advanced Remote Sensing and GIS, 2018
The aim of this study is to assess the land use/land cover (LULC) inter-seasonal changes along th... more The aim of this study is to assess the land use/land cover (LULC) inter-seasonal changes along the Cameroonian shores of Lake Chad and its hinterland using the four generations of Landsat sensors images of MSS, TM, ETM+ and OLI. Identification of land use/land cover inter-seasonal changes is based on classification by Support Vector Machines (SVMs) algorithm. Three major spatial classes of objects are identified: open water, vegetation and marshlands, and bare soils. The results show that, bare soils class has the higher rate, and can reach 67.57% of the study area extent. Moreover, land use/land cover change from one season to another or from one decade to another can be closely linked to evolution of climate conditions. Then open water areas vary little with rate of 1.94% and 7.6% for inter-seasonal changes, and rate of 5.62% and 63.05% for inter-annual changes. Compared to open water, vegetation and marshlands has the most important variation, that is 583.59%. In addition, proportions of bare soils that vary are different between dry seasons and rainy seasons, with a lower and a higher rate of 5.38% and 82.8%. This leads to the conclusion that occupation in the study area is dominated by bare soils principally; occupation class that is most affected by changes is vegetation and marshlands followed by bare soils and then open water and marshlands.
ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2020
Drought affects all human activities and ecosystems. Nearly 40 percent of the world's population ... more Drought affects all human activities and ecosystems. Nearly 40 percent of the world's population inhabit Drylands, and they depend on agriculture for their food, security and livelihoods. Among the remote sensing indices developed, the Land Surface General Drought Index (LSGDI) was recently proposed. This paper proposes an improved model of LSGDI to face the issue of drought in semi-arid and arid regions. The experiment was conducted for the Maga's floodplain, in North-Cameroon. The method uses satellite images of Landsat in 1987, 2003 and 2018, for January and March or April, corresponding to the middle and the end of the dry season. A Vegetation Moisture Index (VMI) and a Normalized Difference Soil Drought Index (NDSoDI) are both developed. On an orthogonal plan, their projections give a drought line that expresses the improved LSGDI (LSGDI2) as the root sum square of the NDSoDI and the VMI. The LSGDI2 results are ranged in [0.09-0.14] interval, which is used to define the threshold and ease the qualifiers for drought classes. The visual patterns easily match the sandy areas of the original Landsat images with the highest values, while the vegetation and water areas match the lowest values. Compared with the LSGDI and Second Modified Perpendicular drought Index (MPDI1), the new index reflectance values are higher. Finally, although LSGDI2 curve's evolution follows the NDSoDI one at 94%, the new spectral index values depends on the both components, helping to map highest values of drought and moisture in Maga's floodplain, for a sustainable rice culture expansion.
International Journal of Advanced Remote Sensing and GIS, 2016
This paper aimed to assess the status of land degradation in arid and semi-arid areas based on a ... more This paper aimed to assess the status of land degradation in arid and semi-arid areas based on a correlation analysis between spectral and statistical neo-bands. The methodology uses vegetation and soil spectral indices as the second Modified Soil Adjusted Vegetation Index (MSAVI2), Normalized Difference Bare Soil Index (NDBSI), Texture Index (NDTeI), Crust Index (CI), Top Soil Grain Size Index (GSI), Normalized Difference Sand Dune Index (NDSDI) and the first Specific Principal Component of the red, near infrared, shortwave infrared bands stacking (SPC1 R-NIR-SWIR1-SWIR2). The vegetation is considered here as the main object of soil sub-surface. Thus after all the spectral and the statistic neo-bands are performed on Landsat8 OLI sensor image, a linear regression is generated to assess their correlation with MSAVI2. Based on the visual interpretation and the regression curves the results show that the determination coefficient R 2 and the P values all significant as less than 0.0001. Each neo-band is weighted with its R 2 to improve its contribution to the model and the synthesis image obtained enhances the land degradation sensing in six classes; these are respectively named as ''severe'' (3139 km 2), ''high'' (6763 km 2), ''moderate'' (8341 km 2), ''low'' (7454 km 2), ''very low'' (6947 km 2) and ''close to nil'' (5437 km 2). This last image is summed with population layer to produce a decision map helpful for further government decision. At the end the degradation image has given interesting results for the detection of land degradation comparatively to derivation and comparison of individual indices.
This chapter proposes a remote sensing multi-angles methodology to assess the transition at the i... more This chapter proposes a remote sensing multi-angles methodology to assess the transition at the interface of the forest-savanna land cover. On Sentinel2-A median images of successive dry seasons, three referential and nine analytical spectral indices were computed. The change vector analysis (CVA) was performed, selecting further one magnitude per index. The averaged moving standard deviation index (aMSDI) was proposed to compare spatial intensity of anomalies among selected CVA, and then statistically assessed through spatial and no-spatial autoregression tests. The cross-correlation and simple linear combination (SCL) computations spotted the overall anomaly extent. Three machine learning algorithms, i.e., classification and regression trees (CART), random forest (RF), and support vector machine (SVM), helped mapping the distribution of each specie. As result, the CVA confirmed each index ability to add new information. The aMSDI gave the harmonized interval [0–0.083] among CVA, c...
This paper explores a spectral vector-based methodology on Landsat 8 bands of the visible wavelen... more This paper explores a spectral vector-based methodology on Landsat 8 bands of the visible wavelengths, that is deep-blue (1) to shortwave infrared (7), to improve the urban land features classification. Using two different ratio models, based on two and three bands' combinations in the cloud environment of Google Earth Engine, the Uncertainty reducing Spectral Vector (USV<sub>r</sub>), the Onward Continuous Spectral Vector (OSV<sub>c</sub>) and the Onward Discontinuous Spectral Vector (OSV<sub>d</sub>) are proposed as new entries for the land use land cover (LULC) classification. Two different sizes of arrays are built, i.e. 42 vectors and 15 vectors corresponding to the same number of derivative bands and new pixels′ values. A decision tree is built in <i>J.48</i> and applied to select the most suitable derivative bands for the analysis. Hereafter, the selected ones are stacked and submitted to five machine learning classifiers using a supervised process, namely, Classification and Regression Trees (CART), Random Forest (RF) Gradient Boosting (GBR), Support Vector Machine (SVM) and Minimum Distance (MD). This method was tested in the two cities of Bamenda and Foumban in west-Cameroon highlands, due to their good representativeness of tropical hilly urban areas' spatial heterogeneity. The results are satisfying for 4/5 classifiers, up to 87% Overall Accuracy, OA, for 0.82 kappa coefficient, KC, in Bamenda, while combining SVM/OSV<sub>d</sub>. Whereas, in Foumban, the classifiers perform up to 85%OA and 0.78 KC for the combination SVM/USV<sub>r</sub>. Only the MD classifier has always performed below 80%OA. The process has been found better than performing classifiers directly on the multispectral (MS) image, by providing more possibilities of hidden spectral indices not yet explored, as far as we know, to detect and discriminate between LULC features, plus an accurate extraction of human settlements.
International Journal of Advanced Remote Sensing and GIS, 2016
This paper aimed to assess the status of land degradation in arid and semi-arid areas based on a ... more This paper aimed to assess the status of land degradation in arid and semi-arid areas based on a correlation analysis between spectral and statistical neo-bands. The methodology uses vegetation and soil spectral indices as the second Modified Soil Adjusted Vegetation Index (MSAVI2), Normalized Difference Bare Soil Index (NDBSI), Texture Index (NDTeI), Crust Index (CI), Top Soil Grain Size Index (GSI), Normalized Difference Sand Dune Index (NDSDI) and the first Specific Principal Component of the red, near infrared, shortwave infrared bands stacking (SPC1R-NIR-SWIR1-SWIR2). The vegetation is considered here as the main object of soil sub-surface. Thus after all the spectral and the statistic neo-bands are performed on Landsat8 OLI sensor image, a linear regression is generated to assess their correlation with MSAVI2. Based on the visual interpretation and the regression curves the results show that the determination coefficient R2 and the P values all significantas less than 0.0001. ...
Oil Palm (Elaeis guineensis Jacq.) has recorded a boom production the last decades and its main p... more Oil Palm (Elaeis guineensis Jacq.) has recorded a boom production the last decades and its main productive zone is inside the tropics that meet the best biophysical conditions. Investors as well as geospatial practitioners are increasingly interested on the best growing and harvesting conditions. So said, the aim of this paper is to select the best oil palm planting site through the best methods combination. The study area is the district of Njimom located in the west-Cameroon, transitional between the equatorial and the climatic zones. In the same GIS environment, the Weighted Linear Combination (WLC) and Fuzzy Analytic Hierarchy Process (FAHP) respectively highlight the subtle differences between capability and suitability, while the Utility Function (UF) helps to assess the consideration of sustainability aspects. The first results consist in eight layers representing natural conditions, that is rainfall, temperatures, sunshine, slope, elevation, soil richness, soil moisture and forest cover, recoded in six classes ranked from 5 to 0 according to the FAO standardised scale. They are crossed using the straightforward method of WLC to give the "Capability layer". The second results consist in three layers related to the social-economical constraints for production, as built-up How to cite this paper: Mfondoum, A.H.N.,
This paper addresses the remote sensing challenging field of urban mixed pixels on a medium spati... more This paper addresses the remote sensing challenging field of urban mixed pixels on a medium spatial resolution satellite data. The tentatively named Normalized Difference Built-up and Surroundings Unmixing Index (NDBSUI) is proposed by using Landsat-8 Operational Land Imager (OLI) bands. It uses the Shortwave Infrared 2 (SWIR2) as the main wavelength, the SWIR1 with the red wavelengths, for the built-up extraction. A ratio is computed based on the normalization process and the application is made on six cities with different urban and environmental characteristics. The built-up of the experimental site of Yaoundé is extracted with an overall accuracy of 95.51% and a kappa coefficient of 0.90. The NDBSUI is validated over five other sites, chosen according to Cameroon's bioclimatic zoning. The results are satisfactory for the cities of Yokadouma and Kumba in the bimodal and monomodal rainfall zones, where overall accuracies are up to 98.9% and 97.5%, with kappa coefficients of 0.88 and 0.94 respectively, although these values are close to those of three other indices. However, in the cities of Foumban, Ngaoundéré and Garoua, representing the western highlands, the high Guinea savannah and the Sudano-sahelian zones where built-up is more confused with soil features, overall accuracies of 97.06%, 95.29% and 74.86%, corresponding to 0.918, 0.89 and 0.42 kappa coefficients were recorded. Difference of accuracy with EBBI, NDBI and UI are up to 31.66%, confirming the NDBSUI efficiency to automate built-up extraction and unmixing from surrounding noises with less biases.
Drought affects all human activities and ecosystems. Nearly 40 percent of the world's population ... more Drought affects all human activities and ecosystems. Nearly 40 percent of the world's population inhabit Drylands, and they depend on agriculture for their food, security and livelihoods. Among the remote sensing indices developed, the Land Surface General Drought Index (LSGDI) was recently proposed. This paper proposes an improved model of LSGDI to face the issue of drought in semi-arid and arid regions. The experiment was conducted for the Maga's floodplain, in North-Cameroon. The method uses satellite images of Landsat, corresponding to the middle and the end of the dry season. A Vegetation Moisture Index (VMI) and a Normalized Difference Soil Drought Index (NDSoDI) are both developed. On an orthogonal plan, their projections give a drought line that expresses the improved LSGDI (LSGDI2) as the root sum square of the NDSoDI and the VMI. The LSGDI2 results are ranged in [0.09-0.14] interval, which is used to define the threshold and ease the qualifiers for drought classes. The visual patterns easily match the sandy areas of the original Landsat images with the highest values, while the vegetation and water areas match the lowest values. Compared with the LSGDI and Second Modified Perpendicular drought Index (MPDI1), the new index reflectance values are higher. Finally, although LSGDI2 curve's evolution follows the NDSoDI one at 94%, the new spectral index values depends on the both components, helping to map highest values of drought and moisture in Maga's floodplain, for a sustainable rice culture expansion.
International Journal of Advanced Remote Sensing and GIS, 2018
The aim of this study is to assess the land use/land cover (LULC) inter-seasonal changes along th... more The aim of this study is to assess the land use/land cover (LULC) inter-seasonal changes along the Cameroonian shores of Lake Chad and its hinterland using the four generations of Landsat sensors images of MSS, TM, ETM+ and OLI. Identification of land use/land cover inter-seasonal changes is based on classification by Support Vector Machines (SVMs) algorithm. Three major spatial classes of objects are identified: open water, vegetation and marshlands, and bare soils. The results show that, bare soils class has the higher rate, and can reach 67.57% of the study area extent. Moreover, land use/land cover change from one season to another or from one decade to another can be closely linked to evolution of climate conditions. Then open water areas vary little with rate of 1.94% and 7.6% for inter-seasonal changes, and rate of 5.62% and 63.05% for inter-annual changes. Compared to open water, vegetation and marshlands has the most important variation, that is 583.59%. In addition, proportions of bare soils that vary are different between dry seasons and rainy seasons, with a lower and a higher rate of 5.38% and 82.8%. This leads to the conclusion that occupation in the study area is dominated by bare soils principally; occupation class that is most affected by changes is vegetation and marshlands followed by bare soils and then open water and marshlands.
Oil Palm (Elaeis guineensis Jacq.) has recorded a boom production the last decades and its main p... more Oil Palm (Elaeis guineensis Jacq.) has recorded a boom production the last decades and its main productive zone is inside the tropics that meet the best biophysical conditions. Investors as well as geospatial practitioners are increasingly interested on the best growing and harvesting conditions. So said, the aim of this paper is to select the best oil palm planting site through the best methods combination. The study area is the district of Njimom located in the west-Cameroon, transitional between the equatorial and the climatic zones. In the same GIS environment, the Weighted Linear Combination (WLC) and Fuzzy Analytic Hierarchy Process (FAHP) respectively highlight the subtle differences between capability and suitability, while the Utility Function (UF) helps to assess the consideration of sustainability aspects. The first results consist in eight layers representing natural conditions, that is rainfall, temperatures, sunshine, slope, elevation, soil richness, soil moisture and forest cover, recoded in six classes ranked from 5 to 0 according to the FAO standardised scale. They are crossed using the straightforward method of WLC to give the "Capability layer". The second results consist in three layers related to the social-economical constraints for production, as built-up area, distance to road and distance to rivers. These layers are recoded in binary with 1 and 0, they are weighted by the FAHP priority vector and membership approach to give the “suitability layer”. Then, the number of occurrences of each aspect of the sustainability is counted in each of the two preceding processes to perform the UF. The resulting value, that is 0.542 for the capability process and 0.315 for the suitability process, serves to weight their respective layers, and their sum gives the final map with the best oil palm site planting in the northern part of the study area, on about 34,950 ha, representing 44.8% of Njimom district.
IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES, 2019
This paper proposes a tool resulting from the merging of the Eisenhower matrix and Analytical Hie... more This paper proposes a tool resulting from the merging of the Eisenhower matrix and Analytical Hierarchy Process to strengthen the prioritization of actions for managers. The case study is from a Ph.D. thesis work, which the first phase is to produce the candidate's schedule. An Eisenhower matrix is designed and priorities are affected to corresponding quadrant, based on the Priority Quotient (PQ) results. As it is too expert, the second phase, concerning a chapter about the calculation of an Accessibility Index is about designing easier readable tool directly usable by the managers to undertake their actions, saving time and investment. A 4*4 matrix called Actions of Governance Matrix (AGM) accompanied by a Partial Priority Quotient (PPQ) scale issued from the AHP process is the resulting tool.
IAETSD JOURNAL FOR ADVANCED RESEARCH IN APPLIED SCIENCES, 2019
The aim of this paper is to assess the impact of Participatory GIS (PGIS) in the process of decis... more The aim of this paper is to assess the impact of Participatory GIS (PGIS) in the process of decision-making in a developing country context. This practice is an alternative process between GIS and Participatory Learning and Action (PLA) methods. It highlights specificities in agriculture and livestocks, health and tourism whose different departments are adjusting to government's efforts to solve local issues while including local populations. The main methods used here are cognitive mapping and focus groups with GIS facilitators. The results obtained are Counting Zones, Health Areas and Touristic Spatial Units delimitations confined or not inside the different district's boundaries, but consensually validated by all the stakeholders. It confirms that the Indigenous System of knowledge (ISK), combined to GIS expertise and administrative coordination, is dynamic in supporting the Cameroon's advances on the field of proximity governance and territorial planning.
Many cities of developing countries sprawl out because there is barely a planning process, genera... more Many cities of developing countries sprawl out because there is barely a planning process, generally with a mixed housing system. Moreover, most local studies mainly use free satellite images of medium to low/coarse resolution, with low accuracy of results. The main goal of this poster was to implement a Geospatial method to assess the metabolism (transformation) of a mid-urban/mid-rural city. Landsat images of 1987, 2003 and 2019 was used. The first point has been to enhance the built-up detection by proposing the Modified New Built-up Index (MNBI), as the root squared of the sum of squares of the New Built-up Index (NBI; Jieli C. et al., 2010) and the Landsat shortwave infrared two (SWIR2) band. Further, the Land Surface Temperature (LST) was computed and a linear regression was made with the MNBI, Bare soil index (Rikimaru A. et al, 2002) and the Normalized Difference Vegetation Index (Rouse, J.W. et al, 1974). Rate of change (RC) and change average (CA) of slopes and intercepts from the linear equation of LST regressed by BI and NDVI , have helped to better understand the urban metabolism. Finally, RC and CA values have been used in a last equation applied to the 2019 LST, to identify and classify areas of mitigation, i.e. replanting.
This poster addresses the issue of drought and desertification in a floodplain close to the Sahar... more This poster addresses the issue of drought and desertification in a floodplain close to the Sahara's desert. The computation of some commonly used remote sensing drought indices as Perpendicular Drought Index (PDI), Normalized Difference Drought Index (NDDI) and Normalized Multi-Bands Drought Index (NMDI), shows a lot of confusion between sands, dry soils and rice crops. So, this poster proposes an alternative method based on Landsat 8 satellite image. The years considered are 1987, 2003 and 2018, and the dry months of january and March/April are used. The first step is the Vegetation Moisture Index (VMI), generated around a NDWI, SAVI and Blue band normalized difference ratio. The second step is the Normalized Difference Soil Drought Index, built on the normalized difference ratio, with adjusting values, between Red and SWIR1 bands on one hand, and blue bands on another. The third step is the Land Surface General Drought Index (LSGDI), a normalized difference composed index between the two previous indices. Finally, an image difference is plotted from the oldest to the recent image index for January and for March/April, the product of the two results are multiplied my an epsilon value (depending on the operator), to obtain the potential desert areas. The results are better than those of the above mentioned indices, and the ongoing experiments on Google Earth Engine are validating the model.
The aim of this poster is to extract the built-up features in a dry desert environment, by applyi... more The aim of this poster is to extract the built-up features in a dry desert environment, by applying remote sensing methods on Sentinel 2 satellite image. The complexity of demarcating built-up spectral signal from sands one is the main challenge. The results obtained show the low rate of spatial external of built up area throughout the study area (0, 56%), the variability in the size of cities and villages, and the differences in organization and distribution of the built up area.
Uploads
Papers by Homere Ngandam
with 1 and 0, they are weighted by the FAHP priority vector and membership approach to give the “suitability layer”. Then, the number of occurrences of each aspect of the sustainability is counted in each of the two preceding processes to perform the UF. The resulting value, that is 0.542 for the capability process and 0.315 for the suitability process, serves to weight their respective layers, and their sum gives the final map with the best oil palm site planting in the northern part of the study area, on about 34,950 ha, representing 44.8% of Njimom district.
with 1 and 0, they are weighted by the FAHP priority vector and membership approach to give the “suitability layer”. Then, the number of occurrences of each aspect of the sustainability is counted in each of the two preceding processes to perform the UF. The resulting value, that is 0.542 for the capability process and 0.315 for the suitability process, serves to weight their respective layers, and their sum gives the final map with the best oil palm site planting in the northern part of the study area, on about 34,950 ha, representing 44.8% of Njimom district.
So, this poster proposes an alternative method based on Landsat 8 satellite image. The years considered are 1987, 2003 and 2018, and the dry months of january and March/April are used. The first step is the Vegetation Moisture Index (VMI), generated around a NDWI, SAVI and Blue band normalized difference ratio. The second step is the Normalized Difference Soil Drought Index, built on the normalized difference ratio, with adjusting values, between Red and SWIR1 bands on one hand, and blue bands on another. The third step is the Land Surface General Drought Index (LSGDI), a normalized difference composed index between the two previous indices. Finally, an image difference is plotted from the oldest to the recent image index for January and for March/April, the product of the two results are multiplied my an epsilon value (depending on the operator), to obtain the potential desert areas.
The results are better than those of the above mentioned indices, and the ongoing experiments on Google Earth Engine are validating the model.