Convolutional Neural Networks

Abstract—CAMTUAL: An interactive mobile app to research fundamental aspects of deep learning technologies to support Smartphones users a 2D-to-3D automatic converter using smartphones built-in cameras. The whole idea inspired by MagicToon App [1]. As 3D video viewing becomes important and virtual reality market started, the request for 3D devices and contents is growing faster. Producing 3D videos, still remains as a big challenge. In this paper we presented a mobile app that uses a 3D deep neural networks algorithm to automatically convert 2D video and images to a stereoscopic 3D format [2]. In comparison to other mobile apps that doesn’t use automatic 2D-to-3D conversion algorithms, our mobile app uses method that trained end-to-end automatically on stereo pairs extracted from existing 3D videos. This novel mobile app approach outperforms baselines in human evaluations and quantitative.

Sign language is a kind of language which is primary used for the deaf and hard of hearing and also used for those who are unable to physically speak. Sign language is only understood by a small percentage of the population. There is a huge communication gap between the deaf community and the hearing majority but it is not acceptable for a nation. Computerized sign language recognition capabilities attempt to break through this communication gap. Due to the advancement of technology, Artificial Intelligence has made human life easier by using various machine learning techniques. Computer vision is one of the most vital fields of artificial intelligence. However, utilizing computer vision to recognize American Sign Language is extremely difficult since sign language is extremely complicated and has a large inter-class variation. In the last few years, Convolutional Neural Network has become an effective way for the classification of multiclass images. In this paper, we have used Convolutional Neural Network (CNN) for the recognition of ASL alphabets. In this study, we have used the Sign Language MNIST dataset which consists of 34627 images where 27,455 training samples and 7172 testing data. The dataset includes 24 alphabets except for the letter J and Z among 26 alphabets. For the recognition of ASL alphabets, at first, we pre-processed our dataset by the normalization technique. After that, we designed a convolutional neural network architecture for extracting features from hand gesture images and then we trained our CNN model through the training dataset. We have finally evaluated our proposed CNN model based on the test dataset and obtained its accuracy to see that how many ASL alphabets recognize correctly. The proposed CNN architecture was able to achieve an accuracy of 99.78% on unseen data.

In recent years, impressive attention has been given for mining the publically available huge amount of data to gain situational awareness, which may help in preventing or decrease the effect of some disaster by taking the correct responses. In this study, an effective Convolutional Neural Networks (CNN) tweet classification system that fully supports the Turkish language has been developed.
In addition, the first-ever Turkish tweet dataset for crisis response is created. This dataset has been carefully preprocessed, annotated, well organized and suitable to be used by all the well-known natural language processing tools. Furthermore, the performance of some well-known machine learning algorithms, i.e., K-Nearest Neighbor (KNN), Naive Bayes (NB), and Support Vector Machine(SVM) was investigated. Then, the performances of the ensemble systems Random Forest (RF), AdaBoost Classifier (AdaBoost), GradientBoosting Classifier (GBC), when used for text (tweets) classification, has been also observed.
A wide range of experiments was performed to investigate the performance of the developed system. As a result, the developed approach has achieved very good performance, robustness, and stability when processing both Turkish and English languages.
Key Words: Crises Management Systems; Tweet Classification; Turkish language; Convolutional Neural Networks; Natural Language Processing.

A series of recent findings strengthened the case for the preventive wearing of masks. Face masks are very important utility in eliminating spread of Covid-19 by blocking droplets. The prevailing logical sentiment says covers are extremely helpful, and even moderately basic home-caused veils to can offer an extraordinary level of insurance against the novel coronavirus. In offices, schools, colleges, industries it is quiet difficult to monitor whether people are wearing masks or not. In such scenarios deep learning could play vital role in identifying whether people are wearing masks or not and identify them. The major aim and objective of this study is to explore the possibility of the computerized detection of mask or spectacles kind of objects and implementing a convolutional neural network using supervised machine learning to reliably detect the presence of mask or spectacle in input images.

A quick and precise crop leaf disease detection is important to increasing agricultural yield in a sustainable manner. We present a comprehensive overview of recent research in the field of crop leaf disease prediction using image processing (IP), machine learning (ML) and deep learning (DL) techniques in this paper. Using these techniques, crop leaf disease prediction made it possible to get notable accuracies. This article presents a survey of research papers that presented the various methodologies, analyzes them in terms of the dataset, number of images, number of classes, algorithms used, convolutional neural networks (CNN) models employed, and overall performance achieved. Then, suggestions are prepared on the most appropriate algorithms to deploy in standard, mobile/embedded systems, drones, robots and unmanned aerial vehicles (UAV). We discussed the performance measures used and listed some of the limitations and future works that requires to be focus on, to extend real time automated crop leaf disease detection system.

One of the most important phase in preventing the spread of the COVID-19 epidemic is to effectively scanning people suspected of being infected. Radiology images of COVID-19 and other lung diseases like bacterial pneumonia, viral pneumonia have common features. X-Ray chest radiography is an effective tool to triage patients with pneumonia without COVID-19 to efficiently allocate hospital resources. However, this similarity makes it difficult for expert radiologists to detect COVID-19 cases. A reliable method for classifying non-COVID-19 and COVID-19 chest X-Ray images could be useful to reduce triage process and diagnose. In this study, COVID-19 is detected from chest X-Ray images using dense convolutional neural network (DenseNet121). In this study, a DenseNet121 model is proposed and experimental studies are performed on an open-access dataset by allocating on different size of test and training. DenseNet121 model can classify chest X-Ray images as COVID-19, normal, and viral pneumonia. Performance evaluations are performed on the accuracy of the techniques, and the results are compared with state-of-the-art deep learning-based approaches such as AlexNet, VGG19, Xception, InceptionV3, and ResNet50. In experimental studies, the highest accuracy (97%) is obtained with the proposed DenseNet121 model.

We are surrounded by sounds, we hear various types of sounds on a day to day basis whether it is music sound, different noises, etc. The urban life is filled with such sounds, which makes it important and highly useful for us to work on these sounds and get some useful information from it so that we can use it efficiently. These sounds are continuously processed by human minds to decipher information about the environment. The same can be done by a machine learning model. It has been seen that convolutional neural networks have been really successful in classifying images, so it becomes a question of interest that how good do they work with different sounds. In this paper, we have worked upon using different deep learning models to see which can be used for the purpose of sound classification. We have used the Urbansound8K dataset which contains 8732 sound excerpts of urban sounds from 10 classes.

There are numerous applications of unmanned aerial vehicles (UAVs) in the management of civil infrastructure assets. A few examples include routine bridge inspections, disaster management, power line surveillance and traffic surveying. As UAV applications become widespread, increased levels of autonomy and independent decision-making are necessary to improve the safety, efficiency, and accuracy of the devices. This paper details the procedure and parameters used for the training of convolutional neural networks (CNNs) on a set of aerial images for efficient and automated object recognition. Potential application areas in the transportation field are also highlighted. The accuracy and reliability of CNNs depend on the network's training and the selection of operational parameters. This paper details the CNN training procedure and parameter selection. The object recognition results show that by selecting a proper set of parameters, a CNN can detect and classify objects with a high level of accuracy (97.5%) and computational efficiency. Furthermore, using a convolutional neural network implemented in the "YOLO" ("You Only Look Once") platform, objects can be tracked, detected ("seen"), and classified ("comprehended") from video feeds supplied by UAVs in real-time.

Contrary to common methods of authorship attribution, by implementing a multi convolutional layer structure, we are able to reduce training time drastically while maintaining relatively high accuracy. Despite all this, the training of the model was able to be completed whilst implementing a relatively low power computer with a single GPU in just a few hours. The network is able to achieve accuracies of 75% in the Victorian Era Novelists Authorship Attribution Dataset.

In this paper, we present an investigation on technical details of the byte-level convolutional layer which replaces the conventional linear word projection layer in the neural language model. In particular, we discuss and compare the effective filter configurations, pooling types and the use of bytes instead of characters. We carry out experiments on language packs released by the IARPA Babel project and measure the performance in terms of perplexity and word error rate. Introducing a convolutional layer consistently improves the results on all languages. Also, there is no degradation from using raw bytes instead of proper Unicode characters, even on syllabic alphabets like Amharic. In addition, we report improvements in word error rate from rescoring lattices and evaluate keyword search performance on several languages.

The aim of image (photo) based emotion recognition system emotion is to detect and classify the emotion from the human face from an image. The automatic system mainly has two components that is extraction of facial feature and classification of facial feature. To classify the feature precisely effective feature classification methods should be applied. Many recent systems use additional information from classification which reduces accuracy as well as increase execution time. In this system Double Local Binary Pattern (DLBP) is used for feature extraction which is a variant of Local Binary Pattern (LBP). Since DLBP has small dimension size which reduces the detection time. To handle local illumination problem Logarithmic Laplace-Domain is proposed (LoL-Domain). Finally, Convolution Neural Network (CNN) is used for an obtained feature for classification.

Optical Character Recognition (OCR) is an extensive research field in image processing and pattern recognition. Traditional character recognition methods cannot distinguish a character or a word from a scanned image. This paper proposes a system, which is to develop a method that uses a personal computer, a portable scanner and an application program that would automatically correct the handwritten answer sheets. For handwritten character recognition, the scanned images are fed through a machine learning classifier known as the Convolutional Neural Network (CNN). Two CNN models were proposed and trained on 250 images that were collected from students at Prince Mohammad Bin Fahd University. The proposed system will finally output the final score of the student by comparing each classified answer with the correct answer. The experimental results exhibited that the proposed system performed a high testing accuracy of 92.86%. The system can be used by the instructors in several educational institutions to automatically grade the handwritten answer sheets of students effectively.

The COVID-19 virus, exactly like in numerous other diseases, can be contaminated from person to person by inhalation. In order to prevent the spread of this virus, which led to a pandemic around the world, a series of rules have been set by governments that people must follow. The obligation to use face masks, especially in public spaces, is one of these rules. Objective: The aim of this study is to determine whether people are wearing the face mask correctly by using deep learning methods. Methods: A dataset consisting of 2000 images was created. In the dataset, images of a person from three different angles were collected in four classes, which are "masked", "non-masked", "masked but nose open", and "masked but under the chin". Using this data, new models are proposed by transferring the learning through AlexNet and VGG16, which are the Convolutional Neural network architectures. Classification layers of these models were removed and, Long-Short Term Memory and Bi-directional Long-Short Term Memory architectures were added instead. Result and conclusions: Although there are four different classes to determine whether the face masks are used correctly, in the six models proposed, high success rates have been achieved. Among all models, the TrVGG16 + BiLSTM model has achieved the highest classification accuracy with 95.67%. Significance: The study has proven that it can take advantage of the proposed models in conjunction with transfer learning to ensure the proper and effective use of the face mask, considering the benefit of society.

Face recognition is one of the important biometric authentication research areas for security purposes in many fields such as pattern recognition and image processing. However, the human face recognitions have the major problem in machine learning and deep learning techniques, since input images vary with poses of people, different lighting conditions, various expressions, ages as well as illumination conditions and it makes the face recognition process poor in accuracy. In the present research, the resolution of the image patches is reduced by the max pooling layer in convolutional neural network (CNN) and also used to make the model robust than other traditional feature extraction technique called local multiple pattern (LMP). The extracted features are fed into the linear collaborative discriminant regression classification (LCDRC) for final face recognition. Due to optimization using CNN in LCDRC, the distance ratio between the classes has maximized and the distance of the featu...

In a factory production line, different industry parts need to be quickly differentiated and sorted for further
process. Parts can be of different colors and shapes. It is tedious for humans to differentiate and sort these
objects in appropriate categories. Automating this process would save more time and cost. In the
automation process, choosing an appropriate model to detect and classify different objects based on
specific features is more challenging. In this paper, three different neural network models are compared to
the object sorting system. They are namely CNN, Fast R-CNN, and Faster R-CNN. These models are
tested, and their performance is analyzed. Moreover, for the object sorting system, an Arduino-controlled 5
DoF (degree of freedom) robot arm is programmed to grab and drop symmetrical objects to the targeted
zone. Objects are categorized into classes based on color, defective and non-defective objects.

In this paper, an initial work of a research is discussed which is to teach young autistic children recognizing human facial expression with the help of computer vision and image processing. This paper mostly discusses the initial work of facial expression recognition using a deep convolutional neural network. The Kaggle's FER2013 dataset has been used to train and experiment with a deep convolutional neural network model. Once a satisfactory result is achieved, the dataset is modified with pictures of four different lighting conditions and each of these datasets is again trained with the same model. This is necessary for the end goal of the research which is to recognize facial expression in any possible environment. Finally, the comparison between results with different datasets is discussed and future work of the project is outlined.

Prostate cancer is the second biggest cause of mortality in men, according to statistics. It's said to be a slow-growing malignancy that doesn't exhibit symptoms until it's advanced. Over the last few years, numerous studies on AI algorithms processing various medical imaging such as CT, MRI, and Ultrasound have been conducted. Using AI to manage prostate cancer would have a significant influence on healthcare. With almost 1.3 million new cases identified each year around the world, cancer experts would have a better grasp of the disease and be able to generate more accurate cancer detection forecasts. We give a review of the usage of CNN applied to several automatic processing tasks of prostate cancer detection and diagnosis, to provide an overview of the progress in this field, based on the increased interest of CNN in recent years. We've noticed that the use of CNN has skyrocketed, with outstanding results obtained either with fresh models or employing pre-trained networks for transfer learning. According to the results of the survey, deep learning-based research outperforms traditional patient prognosis techniques in terms of accuracy.

This paper presents the proposal for the identification of residential equipment in non-intrusive load monitoring systems. The system is based on a Convolutional Neural Network to classify residential equipment. As inputs to the system, transient power signal data obtained at the time an equipment is connected in a residence is used. The methodology was developed using data from a public database (REED) that presents data collected at a low frequency (1 Hz). The results obtained in the test database indicate that the proposed system is able to carry out the identification task, and presented satisfactory results when compared with the results already presented in the literature for the problem in question.

one of the most common malignant tumors in the world today is lung cancer, and it is the primary cause of death from cancer. With the continuous advancement of urbanization and industrialization, the problem of air pollution has become more and more serious. The best treatment period for lung cancer is the early stage. However, the early stage of lung cancer often does not have any clinical symptoms and is difficult to be found. In this paper, lung nodule classification has been performed; the data have used of CT image is SPIE AAPM-Lung. In recent years, deep learning (DL) was a popular approach to the classification process. One of the DL approaches that have used is Transfer Learning (TL) to eliminate training costs from scratch and to train for deep learning with small training data. Nowadays, researchers have been trying various deep learning techniques to improve the efficiency of CAD (computer-aided system) with computed tomography in lung cancer screening. In this work, we implemented pre-trained Convolutional Neural Networks include: AlexNet, ResNet18, Googlenet, and ResNet50 models. These networks are used for training the network and CT image classification. CNN and TL are used to achieve high performance resulting and specify lung cancer detection on CT images. The evaluation of models is calculated by some matrices such as confusion matrix, precision, recall, specificity, and f1-score.

The paper focuses on the problem of raw data
fusion in neural networks based 3D object detection architectures.
Here we consider the case of autonomous driving with data
from camera and LiDAR sensors. Understanding the vehicle
surroundings is a crucial task in autonomous driving since any
subsequent action taken is strongly dependent on it. In this
paper we present an alternative method of fusing camera image
information with LiDAR poinclouds at a close to raw level of
abstraction. Our results suggest that our approach improves
the average precision of 3D bounding box detection of cyclists
(and possibly other objects) in sparse point clouds compared to
the baseline architecture without low-level fusion. The proposed
approach has been evaluated on the KITTI dataset containing
driving scenes with corresponding camera and LiDAR data. The
long-term goal of our research is to develop a neural network
architecture for environment perception that fuses multi-sensor
data at the earliest stages possible, thus leveraging the full benefits
of possible inter-sensor synergies.

The investigation of applying Deep Learning algorithm in order to validate the astronomers' first step which is looking for interesting objects in a binary classification problem of determining what is real and what is not in the underlying massive astronomical data

—The current document blur/quality estimation algorithms rely on the OCR accuracy to measure their success. A sharp document image, however, at times may yield lower OCR accuracy owing to factors independent of blur or quality of capture. The necessity to rely on OCR is mainly due to the difficulty in quantifying the quality otherwise. In this work, we overcome this limitation by proposing a novel dataset for document blur estimation, for which we physically quantify the blur using a capture setup which computationally varies the focal distance of the camera. We also present a selective search mechanism to improve upon the recently successful patch-based learning approaches (using codebooks or convolutional neural networks). We present a thorough analysis of the improved blur estimation pipeline using correlation with OCR accuracy as well as the actual amount of blur. Our experiments demonstrate that our method outperforms the current state-of-the-art by a significant margin.

The current Document Image Quality Assessment (DIQA) algorithms directly relate the Optical Character Recognition (OCR) accuracies with the quality of the document to build supervised learning frameworks. This direct correlation has two major limitations: (a) OCR may be affected by factors independent of the quality of the capture and (b) it cannot account for blur variations within an image. An alternate possibility is to quantify the quality of capture using human judgement, however, it is subjective and prone to error. In this work, we build upon the idea of Spatial Frequency Response (SFR) to reliably quantify the quality of a document image. We present through quantitative and qualitative experiments that the proposed metric leads to significant improvement in document quality prediction in contrast to using OCR as ground truth.

Hand gestures are a natural way for human-robot interaction. Vision based dynamic hand gesture recognition has become a hot research topic due to its various applications. This paper presents a novel deep learning network for hand gesture recognition. The network integrates several well-proved modules together to learn both short-term and long-term features from video inputs and meanwhile avoid intensive computation. To learn short-term features, each video input is segmented into a fixed number of frame groups. A frame is randomly selected from each group and represented as an RGB image as well as an optical flow snapshot. These two entities are fused and fed into a convolutional neural network (ConvNet) for feature extraction. The ConvNets for all groups share parameters. To learn long-term features, outputs from all ConvNets are fed into a long short-term memory (LSTM) network, by which a final classification result is predicted. The new model has been tested with two popular hand gesture datasets, namely the Jester dataset and Nvidia dataset. Comparing with other models, our model produced very competitive results. The robustness of the new model has also been proved with an augmented dataset with enhanced diversity of hand gestures.

The face mask is an essential sanitaryware in daily lives growing during the pandemic period and is a big threat to current face recognition systems. The masks destroy a lot of details in a large area of face, and it makes it difficult to recognize them even for humans. The evaluation report shows the difficulty well when recognizing masked faces. Rapid development and breakthrough of deep learning in the recent past have witnessed most promising results from face recognition algorithms. But they fail to perform far from satisfactory levels in the unconstrained environment during the challenges such as varying lighting conditions, low resolution, facial expressions, pose variation and occlusions. Facial occlusions are considered one of the most intractable problems. Especially when the occlusion occupies a large region of the face because it destroys lots of official features.

The E-Commerce industry is a new way of shopping. Most manufacturers prefer to sell their products on E-commerce websites because of low maintenance and high returns. This paper highlights the application of data science and deep learning methodologies to drive the E-commerce business. The first part of the paper explains the role of data science to identify potential customers, items recommendations and fraud detections. Different stages of building E-commerce websites and the role of proper handling of data is then highlighted. The second part of the paper is about the use of deep learning approaches like Recurrent neural networks in analysing textual information is presented with a demonstration of three architectures. At last, the importance of convolutional neural networks in the E-commerce industry is presented. Fashion related images like dresses, shoes and bags are classified using two separate convolution neural network architectures. The E-commerce industry is changing the shopping experiences of people. They get a lot of choices to choose items from different websites. But the main challenge lies here. There are so many users and items to deal with. Many e-commerce companies have shut down because of bad engagement of customers. With the latest enhancements and cheap availability of computational power, a lot of data is generated [1]. Proper utilization of this data may ultimately lead to the profitability of an E-commerce business. The data generated from an E-commerce website is both structured and unstructured [2]. The structured data includes customer age, location of the customer, set of items bought, rating information of customers on items and Date and time of browsing and purchase. The patterns generated from structured format is not sufficient to understand a person's shopping needs and there is a need to mine useful patterns from unstructured formats. They include reviews of customers on items [3], user click patterns, tweets, social media comments and items viewed by customers. There are many types of E-commerce industries. Amazon, Flipkart, Paytm mall, Myntra, which sell products directly and are called direct E-commerce websites [4]. These items are sold and shipped to the customers as soon as the order is placed. On the other hand, Music streaming websites like Amazon music, Spotify and Video streaming websites like Netflix, Amazon prime video are also a type of E-commerce websites which engage the customers with good recommendations and charge customers on a prepaid basis [5]. Whatever may be the type of E-commerce industry, the ultimate goal is to engage customers and earn good profits. 2. Data Science in E-Commerce In order to use and extract meaningful information from this diversified form of data, data science techniques and methodologies are used. Data science is used in the E-commerce industry in multiple ways. Some of the main aspects are highlighted. 2.1. Market basket analysis This is one of the oldest approaches and it is also known as association analysis in terms of data mining [6]. The main concept is to find the item/items related to items liked/bought by the customer. This is performed by generating frequent association rules. Many parameters like support, confidence, lift, and leverage are used for this. Where support is to find out of all transactions, the number of transactions the item is present. Confidence

This paper describes the implementation of CNN (Convolution Neural Network) based Optical Character Recognition System for Nepali Language, a commonlyspoken language in Nepal. The system has beendeveloped in python using Keras[1] library on top ofTheano[2] and numpy[3]. The system has been trainedusing a set of real world[4] and synthesized data setsconsidering various noise conditions. The tests have alsobeen carried out in a similar setup. The system is furtherenhanced by using nGram based language models. Thispaper details the experiment by discussing the concept,implementation details and overall interpretation of thesystem.

Face recognition technologiesstate of play & technological frontiers Nowadays a simple image of our face could reveal a lot more about us than we would like people to know. This is possible thanks to facial recognition technologies (FRT) that can trace a person's identity and other information if the image of his face can be found in some database, which is very likely, since about 3.81 billion people are active users of social media i. FRT has been studied and tested since the 1960s, however it had failed to break through in the application field as it was unreliable and inaccurate due to technological limitations. Nevertheless, since a little less than a decade 1 , facial recognition has started to be more and more present not only in airports and public open spaces but also implemented in banks' services, computers, smartphones etc. Certainly, most of the credit goes to Artificial Intelligence (AI). The goal of Artificial Intelligence is to emulate human intelligence, and Jeff Mc Frockman defines it as "the ability of machines to make decisions and learn in a similar manner as humans" ii. AI refers to a computer system that can perform tasks that normally require human intelligence such as speech recognition, visual perception, decision making, and language translation. But talking about artificial intelligence applied to facial recognition technologies is rather reductive and generic as it is not possible to understand how this works without talking about Machine learning (ML) and Deep Learning (DL).

Dog breed identification problem has many inter-class similarities and makes it difficult to classify or identify dogs which have similarities, we have about 130 different dog breeds and they are identified by convolutional neural network using transfer learning pre-trained models ResNet50 to obtain an accuracy of 82.7%, we use OpenCV implementation to detect human faces and to predict the nearest resemblance to a breed of dog, so if misclassified as human the predicted breed can be known. We use 30 epochs and a batch size of 64 to achieve this accuracy.

Adding to the literature on the data-driven detection of bid-rigging cartels, we propose a novel approach based on deep learning (a subfield of artificial intelligence) that flags cartel participants based on their pairwise bidding interactions with other firms. More concisely, we combine a so-called convolutional neural network for image recognition with graphs that in a pairwise manner plot the normalized bid values of some reference firm against the normalized bids of any other firms participating in the same tenders as the reference firm. Based on Japanese and Swiss procurement data, we construct such graphs for both collusive and competitive episodes (i.e when a bid-rigging cartel is or is not active) and use a subset of graphs to train the neural network such that it learns distinguishing collusive from competitive bidding patterns. We use the remaining graphs to test the neural network's out-of-sample performance in correctly classifying collusive and competitive bidding ...

— In this paper, we propose a multi-modal search engine for interior design that combines visual and textual queries. The goal of our engine is to retrieve interior objects, e.g. furniture or wall clocks, that share visual and aesthetic similarities with the query. Our search engine allows the user to take a photo of a room and retrieve with a high recall a list of items identical or visually similar to those present in the photo. Additionally, it allows to return other items that aesthetically and stylistically fit well together. To achieve this goal, our system blends the results obtained using textual and visual modalities. Thanks to this blending strategy, we increase the average style similarity score of the retrieved items by 11%. Our work is implemented as a Web-based application and it is planned to be opened to the public.

Alzheimer’s Disease is a severe neurological brain disorder. It destroys brain cells causing people to lose their memory, mental functions and ability to continue daily activities. Alzheimer’s Disease is not curable, but earlier detection can help improve symptoms in a great deal. Machine learning techniques can vastly improve the process for accurate diagnosis of Alzheimer’s Disease. In recent days deep learning techniques have achieved major success in medical image analysis. But relatively little investigation has been done to applying deep learning techniques for Alzheimer’s Disease detection and classification. This paper presents a novel deep learning model for multi-Class Alzheimer’s Disease detection and classification using Brain MRI Data. We design a very deep convolutional network and demonstrate the performance on the Open Access Series of Imaging Studies (OASIS) database.

In general, agriculture is the backbone of India and also plays an important role in Indian economy by providing a certain percentage of domestic product to ensure the food security. But now-a-days, food production and prediction is getting depleted due to unnatural climatic changes, which will adversely affect the economy of farmers by getting a poor yield and also help the farmers to remain less familiar in forecasting the future crops. This research work helps the beginner farmer in such a way to guide them for sowing the reasonable crops by deploying machine learning, one of the advanced technologies in crop prediction. Convolution Neural Network one of the most popular deep neural networks puts forth in the way to achieve it. The soil type image is taken here like alluvial soil, black soil, red soil, sandy soil. Etc. to start the prediction process.

At the conclusion of Our Undemocratic Constitution, Sanford Levinson asks the American people to call a new constitutional convention. Levinson's critics dismissed this call as fanciful, not least because of the assumption that the populace unthinkingly venerates the Constitution too much to countenance the idea of a convention. We challenge the conventional wisdom on conventions by analyzing a 2011 Time magazine poll indicating one in three Americans would support such a call. While constitutional support remains high, we contend the cultural power of law allows citizens to have meaningful and sometimes critical constitutional attitudes. Logistic regression analysis indicates various personal attributes shape these attitudes, including ideology, race, age, income, and constitutional knowledge. Approval of Congress and preferred method of constitutional interpretation also structure convention support.

COVID-19 epidemic has affected our daily life disturbing the world trade and transport. Wearing a face mask has become a new necessity for safety. In the near future, many institutions will ask the customers to wear masks to avail of their services. Therefore, face mask detection has become a necessity to help society. This paper presents a simplified approach to achieve this purpose using some packages like TensorFlow, Keras, OpenCV and Scikit-Learn. This method detects the face from the image in frame and then identifies if it has worn a mask or not. As in a surveillance task, it can also detect a face along with a mask in movement through image processing. The method attains accuracy up to 93% and 91.2% respectively on two datasets. We explore optimized values of parameters using the Sequential CNN (Convolutional Neural Network) model to detect the presence of masks correctly.

In the last decade, market financial forecasting has attracted high interests amongst the researchers in pattern recognition. Usually, the data used for analysing the market, and then gamble on its future trend, are provided as time series; this aspect, along with the high fluctuation of this kind of data, cuts out the use of very efficient classification tools, very popular in the state of the art, like the well known convolutional neural networks (CNNs) models such as Inception, ResNet, AlexNet, and so on. This forces the researchers to train new tools from scratch. Such operations could be very time consuming. This paper exploits an ensemble of CNNs, trained over Gramian angular fields (GAF) images, generated from time series related to the Standard & Poor’s 500 index future; the aim is the prediction of the future trend of the U.S. market. A multi-resolution imaging approach is used to feed each CNN, enabling the analysis of different time intervals for a single observation. A simple trading system based on the ensemble forecaster is used to evaluate the quality of the proposed approach. Our method outperforms the buy-and-hold (B&H) strategy in a time frame where the latter provides excellent returns. Both quantitative and qualitative results are provided.

Nowadays, advances in Artificial Intelligence (AI), especially in machine and deep learning, present new opportunities to build tools that support the work of specialists in areas apparently far from the information technology field. One example of such areas is that of ancient Egyptian hieroglyphic writing. In this study, we explore the ability of different convolutional neural networks (CNNs) to classify pictures of ancient Egyptian hieroglyphs coming from two different datasets of images. Three well-known CNN architectures (ResNet-50, Inception-v3 and Xception) were taken into consideration and trained on the available images. The paradigm of transfer learning was tested as well. In addition, modifying the architecture of one of the previous networks, we developed a specifically dedicated CNN, named Glyphnet, tailoring its complexity to our classification task. Performance comparison tests were performed and Glyphnet showed the best performances in respect to the other CNNs. In conclusion, this work shows how the ancient Egyptian hieroglyphs identification task can be supported by the deep learning paradigm, laying the foundation for information tools supporting automatic documents recognition, classification and, most importantly, the language translation task.

This survey paper encapsulates the various deep learning algorithms employed on video data. Scarce of research has been conducted for classification, clustering and tagging the videos in social networks such as YouTube, Facebook. Instead reading we prefer to watch, hence video huddling is obligatory. The analysis and review of deep learning models for video are explained in this paper. Deep learning is merely in its infancy and, in the epochs to come, will transmute the public. Also, this article discusses the efficiency of deep learning models and concludes with thoughts on future augmentation in video clustering.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

3D convolution neural networks (CNNs) have shown excellent predictive performance on tasks such as action recognition from videos. Since 3D CNNs have unique characteristics and extremely high compute/memory-overheads, executing them on accelerators designed for 2D CNNs provides sub-optimal performance. To overcome these challenges, researchers have recently proposed architectures for 3D CNNs. In this paper, we present a survey of hardware accelerators and hardware-aware algorithmic optimizations for 3D CNNs. We include only those CNNs that perform 3D convolution and not those that perform only 2D convolution on 2D or 3D data. We highlight their key ideas and underscore their similarities and differences. We believe that this survey will spark a great deal of research towards the design of ultra-efficient 3D CNN accelerators of tomorrow.

In each and every field there is a need of maintaining attendance management system. The system will maintain a log details of each and every in their respective field. From past few years Automatic Face Recognition (AFR) technologies have fair improvements in performance, and such systems are now widely used for security and commercial applications. Automated face recognition system is helpful for recognizing face of humans in a real time background mainly for college to mark the attendance of their students. Automatic Face Recognition System is a real world solution which is useful in everyday activities. The matched face is used to mark attendance of the student. This system maintains the attendance records of students. The traditional approach of manually entering of attendance in logbooks is a difficult task and it also wastes the time. So this system is going to design an efficient module that comprises of face recognition to manage the attendance records of students. The new ...