Text Separation From Document Images

Text separation in document image analysis is an important preprocessing step before executing an optical character recognition (OCR) task. It is necessary to improve the accuracy of an OCR system. Traditionally, for separating text from a document, different feature extraction processes have been used that require handcrafting of the features. However, deep learning-based methods are excellent feature extractors that learn features from the training data automatically. Deep learning gives state-of-the-art results on various computer vision, image classification, segmentation, image captioning, object detection, and recognition tasks. This chapter compares various traditional as well as deep-learning techniques and uses a semantic segmentation method for separating text from Devanagari document images using U-Net and ResU-Net models. These models are further fine-tuned for transfer learning to get more precise results. The final results show that deep learning methods give more accurate results compared with conventional methods of image processing for Devanagari text extraction.

Malaria Detection System Using Convolutional Neural Network Algorithm

Malaria is a disease caused when a female Anopheles mosquito bites. There are over 200 million cases recorded per year with more than 400,000 deaths. Current methods of diagnosis are effective; however, they work on technologies that do not produce higher accuracy results. Henceforth, to improve the prediction rate of the disease, modern technologies need to be performed for obtain accurate results. Deep learning algorithms are developed to detect, learn, and determine the containing parasites from the red blood smears. This chapter shows the implementation of a deep learning algorithm to identify the malaria parasites with higher accuracy.

Deep Learning in Engineering Education

The goodness measure of any institute lies in minimising the dropouts and targeting good placements. So, predicting students' performance is very interesting and an important task for educational information systems. Machine learning and deep learning are the emerging areas that truly entice more research practices. This research focuses on applying the deep learning methods to educational data for classification and prediction. The educational data of students from engineering domain with cognitive and non-cognitive parameters is considered. The hybrid model with support vector machine (SVM) and deep belief network (DBN) is devised. The SVM predicts class labels from preprocessed data. These class labels and actual class labels acts as input to the DBN to perform final classification. The hybrid model is further optimised using cuckoo search with Levy flight. The results clearly show that the proposed model SVM-LCDBN gives better performance as compared to simple hybrid model and hybrid model with traditional cuckoo search.

Deep Convolutional Neural Network-Based Analysis for Breast Cancer Histology Images

Breast cancer is one of the main causes of cancer death worldwide, and early diagnostics significantly increases the chances of correct treatment and survival, but this process is tedious. The relevance and potential of automatic classification algorithms using Hematoxylin-Eosin stained histopathological images have already been demonstrated, but the reported results are still sub-optimal for clinical use. Deep learning-based computer-aided diagnosis (CAD) has been gaining popularity for analyzing histopathological images. Based on the predominant cancer type, the goal is to classify images into four categories of normal, benign, in situ carcinoma, and invasive carcinoma. The convolutional neural networks (CNN) is proposed to retrieve information at different scales, including both nuclei and overall tissue organization. This chapter utilizes several deep neural network architectures and gradient boosted trees classifier to classify the histology images among four classes. Hence, this approach has outperformed existing approaches in terms of accuracy and implementation complexity.

Employee's Attrition Prediction Using Machine Learning Approaches

In any industry, attrition is a big problem, whether it is about employee attrition of an organization or customer attrition of an e-commerce site. If we can accurately predict which customer or employee will leave their current company or organization, then it will save much time, effort, and cost of the employer and help them to hire or acquire substitutes in advance, and it would not create a problem in the ongoing progress of an organization. In this chapter, a comparative analysis between various machine learning approaches such as Naïve Bayes, SVM, decision tree, random forest, and logistic regression is presented. The presented result will help us in identifying the behavior of employees who can be attired over the next time. Experimental results reveal that the logistic regression approach can reach up to 86% accuracy over other machine learning approaches.

Deep Learning With PyTorch

In the last decade, deep learning has seen exponential growth due to rise in computational power as a result of graphics processing units (GPUs) and a large amount of data due to the democratization of the internet and smartphones. This chapter aims to throw light on both the theoretical aspects of deep learning and its practical aspects using PyTorch. The chapter primarily discusses new technologies using deep learning and PyTorch in detail. The chapter discusses the advantages of using PyTorch compared to other deep learning libraries. The chapter discusses some of the practical applications like image classification and machine translation. The chapter also discusses the various frameworks built with the help of PyTorch. PyTorch consists of various models that increases its flexibility and accessibility to a greater extent. As a result, many frameworks built on top of PyTorch are discussed in this chapter. The authors believe that this chapter will help readers in getting a better understanding of deep learning making neural networks using PyTorch.

A Novel Deep Learning Method for Identification of Cancer Genes From Gene Expression Dataset

Computational biology is the research area that contributes to the analysis of biological information. The selection of the subset of cancer-related genes is one amongst the foremost promising clinical research of gene expression data. Since a gene can take the role of various biological pathways that in turn can be active only under specific experimental conditions, the stacked denoising auto-encoder(SDAE) and the genetic algorithm were combined to perform biclustering of cancer genes from huge dimensional microarray gene expression data. The Genetic-SDAE proved superior to recently proposed biclustering methods and better to determine the maximum similarity of a set of biclusters of gene expression data with lower MSR and higher gene variance. This work also assesses the results with respect to the discovered genes and spot that the extracted set of biclusters are supported by biological evidence, such as enrichment of gene functions and biological processes.

Emotion Recognition With Facial Expression Using Machine Learning for Social Network and Healthcare

In this chapter, machine learning application on facial expression recognition (FER) is studied for seven emotional states (disgust, joy, surprise, anger, sadness, contempt, and fear) based on FER describing coefficient. FER has many practical importance in various area like social network, robotics, healthcare, etc. Further, a literature review of existing machine learning approaches for FER is discussed, and a novel approach for FER is given for static and dynamic images. Then the results are compared with the other existing approaches. The chapter also covers additional related issues of applications, various challenges, and opportunities in future FER. For security-based face detection systems that can identify an individual, in any form of expression he introduces himself. Doctors will use this system to find the intensity of illness or pain of a deaf and dumb patient. The proposed model is based on machine learning application with three types of prototypes, which are pre-trained model, single layer augmented model, and multi-layered augmented model, having a combined accuracy of approx. 99%.

An Introduction to Deep Convolutional Neural Networks With Keras

Deep learning (DL) is the new buzzword for researchers in the research area of computer vision that unlocked the doors to solving complex problems. With the assistance of Keras library, machine learning (ML)-based DL and various complicated or unresolved issues such as face recognition and voice recognition might be resolved easily. This chapter focuses on the basic concept of Keras-based framework DL library to handle the different real-life problems. The authors discuss the codes of previous libraries and same code run on Keras library and assess the performance on Google Colab Cloud Graphics Processing Units (GPUs). The goal of this chapter is to provide you with the newer concept, algorithm, and technology to solve the real-life problems with the help of Keras framework. Moreover, they discuss how to write the code of standard convolutional neural network (CNN) architectures using Keras libraries. Finally, the codes of validation and training data set to start the training procedure are explored.

Obtaining Deep Learning Models for Automatic Classification of Leukocytes

In this work, the authors classify leukocyte images using the neural network architectures that won the annual ILSVRC competition. The classification of leukocytes is made using pretrained networks and the same networks trained from scratch in order to select the ones that achieve the best performance for the intended task. The categories used are eosinophils, lymphocytes, monocytes, and neutrophils. The analysis of the results takes into account the amount of training required, the regularization techniques used, the training time, and the accuracy in image classification. The best classification results, on the order of 98%, suggest that it is possible, considering a competent preprocessing, to train a network like the DenseNet with 169 or 201 layers, in about 100 epochs, to classify leukocytes in microscopy images.