Deep Convolutional Neural Network for Object Classification

Object classification problem is essential in many applications nowadays. Human can easily classify objects in unconstrained environments easily. Classical classification techniques were far away from human performance. Thus, researchers try to mimic the human visual system till they reached the deep neural networks. This chapter gives a review and analysis in the field of the deep convolutional neural network usage in object classification under constrained and unconstrained environment. The chapter gives a brief review on the classical techniques of object classification and the development of bio-inspired computational models from neuroscience till the creation of deep neural networks. A review is given on the constrained environment issues: the hardware computing resources and memory, the object appearance and background, and the training and processing time. Datasets that are used to test the performance are analyzed according to the images environmental conditions, besides the dataset biasing is discussed.

Apple Leaf Disease Identification Based on Optimized Deep Neural Network

Apples are the most productive fruits in the world with a lot of medicinal and nutritional value. Significant economic losses occur frequently due to various diseases that occur on a huge scale of apple production. Consequently, the effective and timely discovery of apple leaf infection becomes compulsory. The proposed work uses optimal deep neural network for effectively identifying the diseases of apple trees. This work utilizes a convolution neural network to capture the features of Apple leaves. Extracted features are optimized with the help of the optimization algorithm. The optimized features are utilized in the leaf disease identification process. Here the traditional DNN algorithm is modified by means of weight optimization using adaptive monarch butterfly optimization (AMBO) algorithm. The experimental results show that the proposed disease identification methodology based on the optimized deep neural network accomplishes an overall accuracy of 98.42%.

Layer-Wise Tumor Segmentation of Breast Images Using Convolutional Neural Networks

Early prediction of cancer type has become very crucial. Breast cancer is common to women and it leads to life threatening. Several imaging techniques have been suggested for timely detection and treatment of breast cancer. More research findings have been done to accurately detect the breast cancer. Automated whole breast ultrasound (AWBUS) is a new breast imaging technology that can render the entire breast anatomy in 3-D volume. The tissue layers in the breast are segmented and the type of lesion in the breast tissue can be identified which is essential for cancer detection. In this chapter, a u-net convolutional neural network architecture is used to implement the segmentation of breast tissues from AWBUS images into the different layers, that is, epidermis, subcutaneous, and muscular layer. The architecture was trained and tested with the AWBUS dataset images. The performance of the proposed scheme was based on accuracy, loss and the F1 score of the neural network that was calculated for each layer of the breast tissue.

Performance Analysis of GAN Architecture for Effective Facial Expression Synthesis

Facial expressions are an important means of communication among human beings, as they convey different meanings in a variety of contexts. All human facial expressions, whether voluntary or involuntary, are formed as a result of movement of different facial muscles. Despite their variety and complexity, certain expressions are universally recognized as representing specific emotions - for instance, raised eyebrows in combination with an open mouth are associated with surprise, whereas a smiling face is generally interpreted as happy. Deep learning-based implementations of expression synthesis have demonstrated their ability to preserve essential features of input images, which is desirable. However, one limitation of using deep learning networks is that their dependence on data distribution and the quality of images used for training purposes. The variation in performance can be studied by changing the optimizer and loss functions, and their effectiveness is analysed based on the quality of output images obtained.

Pulomonary Nodule Classification From CT Scan Images Using Machine Learning Method

The research work is to enhance the classification accuracy of the pulmonary nodules with the limited number of features extracted using Gray level co-occurrence matrix and linear binary pattern. The classification is done using the machine learning algorithm such as artificial neural network (ANN) and the random forest classifier (RF). In present, lung cancer seems to be the most deadly disease in the world which can be detected only after the computerized tomography (i.e., CT scan images of the person). Detecting the infected portion at the early period is the challenging task. Hence, the recent researchers where under the detection of pulmonary nodules to categorize it either as benign nodules which named as non-cancerous or as malignant nodules which are named as cancerous. When associated the results with the recent papers, the accuracy has been improved in classifying the lung nodules.

Breast Ultrasound Image Processing

Breast cancer is a serious disease among women, and its early detection is very crucial for the treatment of cancer. To assist radiologists who manually delineate the tumour from the ultrasound image an automatic computerized method of detection called CAD (computer-aided diagnosis) is developed to provide valuable inputs for radiologists. The CAD systems is divided into many branches like pre-processing, segmentation, feature extraction, and classification. This chapter solely focuses on the first two branches of the CAD system the pre-processing and segmentation. Ultrasound images acquired depends on the operator expertise and is found to be of low contrast and fuzzy in nature. For the pre-processing branch, a contrast enhancement algorithm based on fuzzy logic is implemented which could help in the efficient delineation of the tumour from ultrasound image.

A New Approach of Deep Learning-Based Tamil Vowels Prediction Using Segmentation and U-Net Architecture

In this chapter, 500 different images of Tamil vowels that are hand written (அஆஇஈஉஊஎஏஐஒஓஔஃ) interprets that the Tamil alphabets model has trained about 75% accuracy with proposed U-net model algorithm. The introduction of various segmentation proportions was discussed for English and Tamil language text identification was explained. In this work, the selection of image is split into four segments and read the data during training itself. Thus, the Tamil and English font prediction accuracy of the model was improved about 85% using U-net architecture was explained.

Performance Analysis of VGG19 Deep Learning Network Based Brain Image Fusion

Multimodal imaging systems assist medical practitioners in cost-effective diagnostic methods in clinical pathologies. Multimodal imaging of the same organ or the region of interest reveals complementing anatomical and functional details. Multimodal image fusion algorithms integrate complementary image details into a composite image that reduces clinician's time for effective diagnosis. Deep learning networks have their role in feature extraction for the fusion of multimodal images. This chapter analyzes the performance of a pre-trained VGG19 deep learning network that extracts features from the base and detail layers of the source images for constructing a weight map to fuse the source image details. Maximum and averaging fusion rules are adopted for base layer fusion. The performance of the fusion algorithm for multimodal medical image fusion is analyzed by peak signal to noise ratio, structural similarity index, fusion factor, and figure of merit. Performance analysis of the fusion algorithms is also carried out for the source images with the presence of impulse and Gaussian noise.

Breast Cancer Detection Using Random Forest Classifier

Breast cancer is the second most prevalent type of cancer among women. Breast ultrasound (BUS) imaging is one of the most frequently used diagnostic tools to detect and classify abnormalities in the breast. To improve the diagnostic accuracy, computer-aided diagnosis (CAD) system is helpful for breast cancer detection and classification. Normally, a CAD system consists of four stages: pre-processing, segmentation, feature extraction, and classification. In this chapter, the pre-processing step includes speckle noise removal using speckle reducing anisotropic diffusion (SRAD) filter. The goal of segmentation is to locate the region of interest (ROI) and active contour-based segmentation and fuzzy C means segmentation (FCM) are used in this work. The texture features are extracted and fed to a classifier to categorize the images as normal, benign, and malignant. In this work, three classifiers, namely k-nearest neighbors (KNN) algorithm, decision tree algorithm, and random forest classifier, are used and the performance is compared based on the accuracy of classification.

Computer Processing of an Image

In this chapter, the authors have briefed about images, digital images, how the digital images can be processed. Image types like binary image, grayscale image, color image, and indexed image and various image formats are explained. It highlights the various fields where digital image processing can be used. This chapter introduces a variety of concepts related to digital image formation in a human eye. The mechanism of the human visual system is discussed. The authors illustrate the steps of image processing. Explanation on different elements of digital image processing systems like image acquisition, and others are also provided. The components required for capturing and processing the image are discussed. Concepts of image sampling, quantization, image representation are discussed. It portrays the operations of the image during sampling and quantization and the two operations of sampling which is oversampling and under-sampling. Readers can appreciate the key difference between oversampling and under-sampling applied to digital images.