Chapter 2 - Convolution neural network and deep learning

Chapter 4 - Deep Learning in Texture Analysis and Its Application to Tissue Image Classification

Convolutional neural networks (CNN) applications range from recognition and reasoning (such as handwriting recognition, facial expression recognition and video surveillance) to intelligent text applications such as semantic text analysis and natural language processing applications. Two key observations drive the design of a new architecture for CNN. First, CNN workloads exhibit a widely varying mix of three types of parallelism: parallelism within a convolution operation, intra-output parallelism where multiple input sources (features) are combined to create a single output, and inter-output parallelism where multiple, independent outputs (features) are computed simultaneously. Workloads differ significantly across different CNN applications, and across different layers of a CNN. Second, the number of processing elements in an architecture continues to scale (as per Moore's law) much faster than off-chip memory bandwidth (or pin-count) of chips. Based on these two observations, we show that for a given number of processing elements and off-chip memory bandwidth, a new CNN hardware architecture that dynamically configures the hardware on-the-fly to match the specific mix of parallelism in a given workload gives the best throughput performance. Our CNN compiler automatically translates high abstraction network specification into a parallel microprogram (a sequence of low-level VLIW instructions) that is mapped, scheduled and executed by the coprocessor. Compared to a 2.3 GHz quad-core, dual socket Intel Xeon, 1.35 GHz C870 GPU, and a 200 MHz FPGA implementation, our 120 MHz dynamically configurable architecture is 4x to 8x faster. This is the first CNN architecture to achieve real-time video stream processing (25 to 30 frames per second) on a wide range of object detection and recognition tasks.

Recognition of human activity based on convolutional neural network has received the interest of researchers in recent years due to its significant improvement in accuracy. A large number of algorithms based on the deep learning approach have been proposed for activity recognition purpose. However, with the increasing advancements in technologies having limited computational resources, it needs to design an efficient deep learning-based approaches with improved utilization of computational resources. This paper presents a simple and efficient 2-Dimensional convolutional neural network (2-D CNN) architecture with very small size convolutional kernel for human activity recognition. The merit of the proposed CNN architecture over standard deep learning architectures is fewer trainable parameters and lesser memory requirement which enables it to train the proposed CNN architecture on low GPU memory-based devices and also works well with smaller as well as larger size datasets. The proposed approach consists of mainly four stages: namely (1) creation of dataset and data augmentation, (2) designing 2-D convolutional neural network (CNN) architecture, (3) the proposed 2-D CNN architecture trained from scratch up to optimum stage, and (4) evaluation of the trained 2-D CNN architecture. To illustrate the effectiveness of the proposed architecture several extensive experiments are conducted on three publicly available datasets, namely IXMAS, YouTube, and UCF101 dataset. The results of the proposed method and its comparison with other state-of-the-art methods [8-12,14,18-26,29-33] demonstrate the usefulness of the proposed method.

Covid-19 first occurred in Wuhan, China in December 2019. Subsequently, the virus spread throughout the world and as of June 2020 the total number of confirmed cases are above 4.7 million with over 315,000 deaths. Machine learning algorithms built on radiography images can be used as a decision support mechanism to aid radiologists to speed up the diagnostic process. The aim of this work is to conduct a critical analysis to investigate the applicability of convolutional neural networks (CNNs) for the purpose of COVID-19 detection in chest X-ray images and highlight the issues of using CNN directly on the whole image. To accomplish this task, we use 12-off-the-shelf CNN architectures in transfer learning mode on 3 publicly available chest X-ray databases together with proposing a shallow CNN architecture in which we train it from scratch. Chest X-ray images are fed into CNN models without any preprocessing to replicate researches used chest X-rays in this manner. Then a qualitative investigation performed to inspect the decisions made by CNNs using a technique known as class activation maps (CAM). Using CAMs, one can map the activations contributed to the decision of CNNs back to the original image to visualize the most discriminating region(s) on the input image. We conclude that CNN decisions should not be taken into consideration, despite their high classification accuracy, until clinicians can visually inspect and approve the region(s) of the input image used by CNNs that lead to its prediction.

An experimental study on classification of thyroid histopathology images using transfer learning

With the continuous development of deep learning technology, neural networks such as convolutional neural network (CNN) have shown good performance in many fields, such as image processing. Meanwhile, the relevant algorithm has made great progress. But the experiment results show that the deeper the network layers, the more the number of parameters that need to be trained in neural network, and the massive computing resources will be consumed to reconstruct and train the deep convolutional neural network (DCNN) model. These parameters often need to be trained in large dataset. But in many practical applications, the effective sample dataset that can be collected are usually small and lack of annotated samples. It is a pity that models that perform well on large datasets often have overfitting problems when applied to small datasets. And transfer learning can recognize and apply knowledge and skills learned in previous domains/tasks to novel domains/tasks. Combining deep convolutional neural network learning with transfer learning can make full use of existing models with good performance to solve problems in new fields, so has received considerable attentions due to its high research value and wide application prospect. This paper focuses on the combination of CNN and transfer learning, analyzes their characteristics, summarizes the relevant models, methods and applications, so as to promote their effective fusion in image classification.

Deep Learning (DL) is a fastest growing and a broader part of machine learning family. Deep learning uses Convolutional Neural Networks (CNN) for image classification as it gives the most accurate results in solving real- world problem. CNN has various pre-trained architecture like AlexNet, GoogleNet, DenseNet, SqueezeNet, ResNet, VGGNet etc. In this study, we have used CNN and AlexNet architecture for detecting the disease in Mango and Potato leaf and compare the accuracy and efficiency between these architectures. The dataset containing 4004 images were used for this work. The images for potato were taken from plantvillage website, while images for mango were collected from GBPUAT field location. The results show that accuracy achieved from AlexNet is higher than CNN architecture.

This paper demonstrates image classification using deep learning. Deep learning has the inherent ability to automatically discover and extract meaningful features for specific applications. Among the popular techniques in deep learning, the convolutional neural network (CNN) stands out. CNN consists of an input layer, hidden layers, and an output layer, where meaningful features are automatically extracted from input images. This paper presents the performance and identifies the most effective CNN architectures for accurately classifying images in the CIFAR-10 dataset. Five CNN architectures were implemented, namely [Architecture 1], [Architecture 2], [Architecture 3], [Architecture 4], and [Architecture 5] using the CIFAR-10 dataset. The architectures were selected based on the need to explore variations in convolutional filter sizes, dense layers, and batch normalization to assess their impact on CIFAR-10 image classification performance. Each architecture was trained on a standard training set and evaluated on a validation set. We used specific details on data preprocessing and training settings for a consistent and fair comparison. After training and evaluation, we have obtained the following results for each architecture. Architecture 1 has a training accuracy of 74.7% and validation accuracy of 76.6%, Architecture 2 has 96.09% and 86.08%, Architecture 3 has 77% and 78.1%, Architecture 4 has 67.91% and 69.54%, and Architecture 5 has 94.64% and 87.34% of training accuracy and validation accuracy respectively. After conducting a comparative analysis, we found that Architecture 5 has achieved the highest validation accuracy in classifying images in the CIFAR-10 dataset. These findings suggest that Architecture 5 is a promising choice for image classification tasks involving the CIFAR-10 dataset.

Medical image classification technology, preferably which is based on the deep learning, is not only a key auxiliary diagnosis and treatment method in clinical medicine but also an important direction of scientific research. With the intensification of social aging, the incidence of viral elderly pneumonia has been on the rise and needs dedication from the research community. Doctors rely on personal theories and experience to use traditional methods to check the computed tomography (CT) images of the lungs of elderly patients one by one, which is likely to cause diagnosis errors. The accuracy of the traditional method certainly meets the clinical needs, but it has higher requirements on the theory and experience of medical staff, and the classification efficiency is low. Constructing an accurate and fast auxiliary system can effectively save medical resources. In response to the above problems, we have proposed a viral pneumonia diagnosis method for lung CT images, which is based on the convolutional neural networks. The main research work is carried out around the following aspects: First, in the lung CT image classification task, the traditional methods are inefficient and effective for doctors. The basic quality requirements of the model are high, or, in the model training, the effective training data are small, and so forth, causing problems such as model overfitting. A lung CT classification model based on the improved Inception-ResNet is proposed. In this model, first the overall architecture of the network model is designed, and then the Contrast-Limited Adaptive Histogram Equalization (CLAHE) algorithm is used to perform the same image enhancement processing on the dataset and data needed in this article, and then the pictures pass through three different network models. A binary classification study was carried out on viral pneumonia and normal lung images, and finally the accuracy, sensitivity, and specificity of the three models were compared. The experimental results show that the accuracy of the three models for the judgment of viral pneumonia is more than 95%. Among these, the model proposed in this article has better classification effect and fit, the highest accuracy rate, and less parameters and can be used for rapid screening of viral senile pneumonia. Objective. To complete the classification of lung CT images of the elderly with viral pneumonia based on the improved Inception-ResNet network architecture. Methods. (1) Find and study domestic and foreign medical literature, understand the diagnosis and treatment methods of viral pneumonia, and study lung CT imaging; compare the pattern classifications of deep learning in lung imaging at home and abroad, and further study the application of convolutional neural networks in the medical field application. (2) Study various models and technologies of convolutional neural networks, summarize them separately, and have in-depth understanding of convolutional neural networks, including architecture, methods, and related system frameworks, experimental environments, and so forth. Results. This paper proposes an optimized Inception-ResNet network architecture for image classification. The control experimental model uses two network models, GoogLeNet and ResNet, and selects the viral pneumonia dataset for training and testing. The experimental results are as follows: the sensitivity and specificity are superior to those in the other two models, which can be used for actual medical screening and diagnosis. Conclusion. The improved Inception-ResNet network model method in this paper performs better in terms of accuracy, sensitivity, and specificity. Every metric is higher than those in the ResNet model and the GoogLeNet model, improving the classification effect. In addition, it can be seen from the experimental results that the model used in this paper has a very good classification effect in the classification of new coronary pneumonia CT image data.

Convolutional Neural Networks (CNNs) have become foundational in addressing image-based tasks across diverse domains such as medical imaging, autonomous driving, remote sensing, and surveillance. Recent developments in CNN architectures have significantly enhanced model performance, efficiency, and generalization capabilities. Innovations such as residual connections, depthwise separable convolutions, attention mechanisms, and hybrid models combining CNNs with transformers have led to breakthroughs in image classification, object detection, image segmentation, and super-resolution. In addition, advances in training strategies, including self-supervised learning and transfer learning, have expanded the applicability of CNNs to scenarios with limited labeled data. This paper reviews and synthesizes the latest advancements in CNN architectures and training techniques, analyzes their performance across key benchmarks, and discusses ongoing challenges and future research directions in the field of image-based AI systems.

In this paper, we study the performance of variants of well-known Convolutional Neural Network (CNN) architectures on different audio tasks. We show that tuning the Receptive Field (RF) of CNNs is crucial to their generalization. An insufficient RF limits the CNN's ability to fit the training data. In contrast, CNNs with an excessive RF tend to over-fit the training data and fail to generalize to unseen testing data. As state-of-the-art CNN architectures - in computer vision and other domains - tend to go deeper in terms of number of layers, their RF size increases and therefore they degrade in performance in several audio classification and tagging tasks. We study well-known CNN architectures and how their building blocks affect their receptive field. We propose several systematic approaches to control the RF of CNNs and systematically test the resulting architectures on different audio classification and tagging tasks and datasets. The experiments show that regularizing the RF of CNNs using our proposed approaches can drastically improve the generalization of models, out-performing complex architectures and pre-trained models on larger datasets. The proposed CNNs achieve state-of-the-art results in multiple tasks, from acoustic scene classification to emotion and theme detection in music to instrument recognition, as demonstrated by top ranks in several pertinent challenges (DCASE, MediaEval).

With the rapid development of convolutional neural networks (CNN) in the field of deep learning, there has been a great process in brain-computer interaction (BCI) in the past few years due to its efficient processing of images. The review aims to discuss the application of CNN in the BCI design module, especially the data processing module, and analyses the problems and corresponding solutions in the researchers process. The author first briefly describes the application of the CNN architecture in BCI design modules and reviews various research in recent years on regarding improving the accuracy of CNN in BCI data processing modules and maintain high performance, including research on models that tries to solve the problem of commands, changes in feature extraction methods, computational methods, and changing the dimensions of the CNN architecture to attain better computation, etc. By analyzing the problems caused by some researchers using CNN architecture in BCI deep learning applications and related solutions, the author summarizes some of the shortcomings of CNN in BCI research and proposes possible research directions.

The advancements in deep learning (DL) and transfer learning (TL) have transformed artificial intelligence, particularly in image classification. This research examines the theoretical foundations of DL and TL, focusing on their applications in medical image classification, specifically distinguishing between COVID-19, viral pneumonia, and normal lung conditions. By leveraging GPU-enabled high-performance computing and large labeled datasets, DL models particularly Convolutional Neural Networks (CNNs) such as ResNet50 and VGG16 have achieved superior accuracy compared to traditional machine learning methods. This study explores feature extraction using pre-trained models, the implementation of classifiers like Support Vector Machines (SVM) and K-Nearest Neighbors (KNN), and the integration of multi-view learning techniques, such as early fusion. The results demonstrate the effectiveness of DL and TL in improving classification performance, highlighting their significant potential to advance global healthcare diagnostics.

In this paper, we propose to address the downscaling of ocean remote sensing data using image super-resolution models based on deep learning, and more particularly Convolutional Neural Networks (CNNs). The goal of this study, for which we focus on satellite-derived Sea Surface Temperature (SST) data, is to evaluate the efficiency and the relevance of deep learning architectures applied to oceanographic remote sensing data. By using a CNN architecture, namely SRCNN (Super Resolution CNN), on a large-scale dataset of SST fields, we show that it allows a considerable gain in terms of PSNR compared to classical downscaling techniques. These results point out the relevance of deep learning models specifically trained for ocean remote sensing data and advocate for other applications to the reconstruction of high-resolution sea surface geophysical fields from multi-sensor satellite observations.

EEG based multi-class seizure type classification using convolutional neural network and transfer learning