College of Computer & Information Sciences

Pedestrian path prediction is an emerging topic in the crowd visual analysis domain, notwithstanding its practical importance in many respects. To date, the few contributions in the literature proposed quite straightforward approaches, and only a few of them have taken into account the interaction between pedestrians as a paramount cue in forecasting their potential walking preferences in a given scene. Moreover, the typical trend was to evaluate the proposed algorithms on sparse scenarios. To cope with more realistic cases, in this paper, we present an efficient approach for pedestrian path prediction in densely crowded scenes. The proposed approach initiates by extracting motion features related to the target pedestrian and his/her neighbors. Second, in order to further increase the representativeness of the extracted motion cues, an autoencoder feature learning model is considered, whose outcome finally feeds a Gaussian process regression prediction model to infer the potential future trajectories of the target pedestrians given their walking records in the scene. Experimental results demonstrate that our framework scores plausible results and outperforms traditional methods in the literature. INDEX TERMS Crowd analysis, walking path prediction, motion modeling, computer vision.

Ear print is an imminent biometric modality that has been attracting increasing attention in the biometric community. However, compared with well-established modalities, such as face and fingerprints, a limited number of contributions has been offered on ear imaging. Moreover, only several studies address the aspect of ear characterization (i.e., feature design). In this respect, in this paper, we propose a novel descriptor for ear recognition. The proposed descriptor, namely, dense local phase quantization (DLPQ) is based on the phase responses, which is generated using the well-known LPQ descriptor. Furthermore, local dense histograms are extracted from the horizontal stripes of the phase maps followed by a pooling operation to address viewpoint changes and, finally, concatenated into an ear descriptor. Although the proposed DLPQ descriptor is built on the traditional LPQ, we particularly show that drastic improvements (of over 20%) are attained with respect to this latter descriptor on two benchmark data sets. Furthermore, the proposed descriptor stands out among recent ear descriptors from the literature. INDEX TERMS Ear imaging, biometrics, ear recognition, feature design, local histograms.

In this article, we propose a novel approach based on convolutional features and sparse autoencoder (AE) for scene-level land-use (LU) classification. This approach starts by generating an initial feature representation of the scenes under analysis from a deep convolutional neural network (CNN) pre-learned on a large amount of labelled data from an auxiliary domain. Then these convolutional features are fed as input to a sparse AE for learning a new suitable representation in an unsupervised manner. After this pre-training phase, we propose two different scenarios for building the classification system. In the first scenario, we add a softmax layer on the top of the AE encoding layer and then fine-tune the resulting network in a supervised manner using the target training images available at hand. Then we classify the test images based on the posterior probabilities provided by the softmax layer. In the second scenario, we view the classification problem from a reconstruction perspe...

Minimizing data overheads in packet networks is an essential performance promotion issue. One of the past solutions was to select a single optimum packet size that minimizes the combined data overhead factor resulting from both operational and blank padding requirements. Such a packet size is associated with the characteristics of the random streams of messages applied to the network. This paper provides a different method that lead to improved minimization of the overheads. The method is based on selecting multiple packet sizes, with each being concerned with minimizing data overheads of messages falling within a certain range of length. The performance of the method was studied using computer simulation. The results obtained illustrate that in comparison with the old method, the new method can reduce the combined data overhead factor by approximately 25%. Future use of the method would provide more efficient data flow through packet networks.

COVID-19 has evolved into one of the most severe and acute illnesses. The number of deaths continues to climb despite the development of vaccines and new strains of the virus have appeared. The early and precise recognition of COVID-19 are key in viably treating patients and containing the pandemic on the whole. Deep learning technology has been shown to be a significant tool in diagnosing COVID-19 and in assisting radiologists to detect anomalies and numerous diseases during this epidemic. This research seeks to provide an overview of novel deep learning-based applications for medical imaging modalities, computer tomography (CT) and chest X-rays (CXR), for the detection and classification COVID-19. First, we give an overview of the taxonomy of medical imaging and present a summary of types of deep learning (DL) methods. Then, utilizing deep learning techniques, we present an overview of systems created for COVID-19 detection and classification. We also give a rundown of the most we...

Searching images from the large image databases is one of the potential research areas of multimedia research. The most challenging task for nay CBIR system is to capture the high level semantic of user. The researchers of multimedia domain are trying to fix this issue with the help of Relevance Feedback (RF). However existing RF based approaches needs a number of iteration to fulfill user's requirements. This paper proposed a novel methodology to achieve better results in early iteration to reduce the user interaction with the system. In previous research work it is reported that SVM based RF approach generating better results for CBIR. Therefore, this paper focused on SVM based RF approach. To enhance the performance of SVM based RF approach this research work applied Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) before applying SVM on user feedback. The main objective of using these meta-heuristic was to increase the positive image sample size from SVM. Firstly steps PSO is applied by incorporating the user feedback and secondly GA is applied on the result generated through PSO, finally SVM is applied using the positive sample generated through GA. The proposed technique is named as Particle Swarm Optimization Genetic Algorithm-Support Vector Machine Relevance Feedback (PSO-G A-SVM-RF). Precisions, recall and F-score are used as performance metrics for the assessment and validation of PSO-GA-SVM-RF approach and experiments are conducted on coral image dataset having 10908 images. From experimental results it is proved that PSO-GA-SVM-RF approach outperformed then various well known CBIR approaches.

Liver cancer is a life-threatening illness and one of the fastest-growing cancer types in the world. Consequently, the early detection of liver cancer leads to lower mortality rates. This work aims to build a model that will help clinicians determine the type of tumor when it occurs within the liver region by analyzing images of tissue taken from a biopsy of this tumor. Working within this stage requires effort, time, and accumulated experience that must be possessed by a tissue expert to determine whether this tumor is malignant and needs treatment. Thus, a histology expert can make use of this model to obtain an initial diagnosis. This study aims to propose a deep learning model using convolutional neural networks (CNNs), which are able to transfer knowledge from pre-trained global models and decant this knowledge into a single model to help diagnose liver tumors from CT scans. Thus, we obtained a hybrid model capable of detecting CT images of a biopsy of a liver tumor. The best r...

This paper deals with the problem of the classification of large-scale very high-resolution (VHR) remote sensing (RS) images in a semisupervised scenario, where we have a limited training set (less than ten training samples per class). Typical pixel-based classification methods are unfeasible for large-scale VHR images. Thus, as a practical and efficient solution, we propose to subdivide the large image into a grid of tiles and then classify the tiles instead of classifying pixels. Our proposed method uses the power of a pretrained convolutional neural network (CNN) to first extract descriptive features from each tile. Next, a neural network classifier (composed of 2 fully connected layers) is trained in a semisupervised fashion and used to classify all remaining tiles in the image. This basically presents a coarse classification of the image, which is sufficient for many RS application. The second contribution deals with the employment of the semisupervised learning to improve the ...

In this paper, we propose a new payment instrument, i.e., mobile traveller's check (MTC) in the realm of mobile commerce. This payment instrument provides the merits of both e-cash and e-check, i.e., MTC can be used freely as an e-cash and it is as secure as an e-check. We present the mobile payment protocol based on MTC which uses elliptic curve digital signature algorithm (ECDSA) for generating and verifying digital signatures and DES for encrypting and decrypting the messages which are suitable for resource constrained devices like mobile phones. We use 'extended BAN' logic (Abadi et al., 1993) to provide a concise and clear understanding of this secure payment instrument (MTC). We formalise and verify the interactions and trust relationships among engaging entities.

The automatic detection of vocal cord/vocal tract diseases from the medically disordered speech is a challenging task in the medical field. An automatic diagnostic system for the medically disordered voice is proposed in this paper. The developed system takes speech recorded by a patient having a disorder in his/her voice, as an input. The system gives the type of voice disorder as an output. The output is accompanied by a confidence score that tells about the systems confidence of correctly recognizing the type of the disorder as output. In the system, after the detection of the vowels from the disordered speech, two formants F1 and F2 are extracted from the vowels and Euclidean distance is used for classification. The results of the proposed system are encouraging.

A new feature called multi-directional local feature (MDLF) is proposed and applied in automatic speaker recognition. In order to extract the MDLF, a windowed speech signal is processed by fast Fourier transform and passed through 24 mel-scaled Filter Bank, followed by log compression. A three-point linear regression is then applied in four different directions, which are horizontal (time axis), vertical (frequency axis), 45 • (time-frequency) and 135 • (time-frequency). MDLF holds the characteristics of the speaker in time spectrum and results in better performance. In the experiments conducted, a Gaussian mixture model (GMM) with a different number of mixtures is used as the classifier. Experimental results show that the proposed MDLF has better recognition accuracy than the traditional MFCC features. The MDLF achieves excellent results both in text-dependent and text-independent speaker recognition, and in Arabic and English speech. The proposed technique is also language independent.

This paper investigates the gender effect in speaker trait recognition especially in likability and personality detection. The acoustic features, classification methods, and feature selection techniques are adopted from the prescribed platform of the Interspeech 2012 Speaker Trait Challenge. In the proposed method, first we separate the files according to gender. Then features and classifiers are applied on gender dependent cases. In the experiments, we find that gender dependent trait recognition is higher than gender independent cases. We also find that the features and classification methods for male and female are different from the best cases. Our proposed technique outperforms the baseline result provided in the challenge in both likability and personality detection.

A new speech feature extraction technique called moving average multi directional local features (MA-MDLF) is presented in this paper. This method is based on linear regression (LR) and moving average (MA) in the time-frequency plane. Three-point LR is taken along time axis and frequency axis, and 3 points MA is taken along 45°and 135°in the timefrequency plane. The LR captures the voice onset\offset, formant contour, while the moving average captures the dynamics on time-frequency axes which can be seen as voiceprints. The MA-MDLF performance is compared to commonly used speech features in speaker recognition. The comparison is performed in a speaker recognition system (SRS) for three different conditions, namely clean speech, mobile speech, and cross channel. MA-MDLF has shown better performance than the baseline MFCC, RASTA-PLP and LPCC. In clean and mobile speech, MA-MDLF feature performs the best and also in the cross channel task MA-MDLF performed excellent. We also evaluated the MA-MDLF using three speech databases, namely KSU, LDC Babylon and TIMITdatabases, and found that MA-MDLF outperformed the other commonly used features with speech from all the three databases. The first and second databases are for Arabic speech while third is for English speech.

Facial expression recognition plays an important role in communicating the emotions and intentions of human beings. Facial expression recognition in uncontrolled environment is more difficult as compared to that in controlled environment due to change in occlusion, illumination, and noise. In this paper, we present a new framework for effective facial expression recognition from real-time facial images. Unlike other methods which spend much time by dividing the image into blocks or whole face image, our method extracts the discriminative feature from salient face regions and then combine with texture and orientation features for better representation. Furthermore, we reduce the data dimension by selecting the highly discriminative features. The proposed framework is capable of providing high recognition accuracy rate even in the presence of occlusions, illumination, and noise. To show the robustness of the proposed framework, we used three publicly available challenging datasets. Th...

There have been several empirical studies addressing breast cancer using machine learning and soft computing techniques. Many claim that their algorithms are faster, easier, or more accurate than others are. This study is based on genetic programming and machine learning algorithms that aim to construct a system to accurately differentiate between benign and malignant breast tumors. The aim of this study was to optimize the learning algorithm. In this context, we applied the genetic programming technique to select the best features and perfect parameter values of the machine learning classifiers. The performance of the proposed method was based on sensitivity, specificity, precision, accuracy, and the roc curves. The present study proves that genetic programming can automatically find the best model by combining feature preprocessing methods and classifier algorithms.

Contrast enhancement is important and plays vital role in many applications. Histogram equalization-based techniques are widely used techniques for contrast enhancement. However, it faces the contrast over-stretching, which in return causes the loss of details and unnatural look to the target image. To address this issue, this work presents a novel scheme for image contrast enhancement. The contribution of the proposed scheme is twofold. First, the image can lose many important information when an image size is decreased. For that, the image is transformed from spatial to wavelet domain so that the multi-resolution can be achieved. Second, Gamma correction is a proven technique that produces natural look and preserves mean brightness of an image with the choice of optimal gamma values. Here, Particle Swarm Optimization (PSO) is utilized to select the optimal gamma values. In this study, an effective fitness function is proposed to maximize the performance of PSO. Experimental findings show that the proposed approach improve the image contrast up to a greater extent without introducing any artifacts.