Traditional systems of handwriting recognition have relied on handcrafted features and a large amount of prior knowledge. Training an Optical character recognition (OCR) system based on these prerequisites is a challenging task. Research in the handwriting recognition field is focused around deep learning techniques and has achieved breakthrough performance in the last few years. Still, the rapid growth in the amount of handwritten data and the availability of massive processing power demands improvement in recognition accuracy and deserves further investigation. Convolutional neural networks (CNNs) are very effective in perceiving the structure of handwritten characters/words in ways that help in automatic extraction of distinct features and make CNN the most suitable approach for solving handwriting recognition problems. Our aim in the proposed work is to explore the various design options like number of layers, stride size, receptive field, kernel size, padding and dilution for CNN-based handwritten digit recognition. In addition, we aim to evaluate various SGD optimization algorithms in improving the performance of handwritten digit recognition. A network’s recognition accuracy increases by incorporating ensemble architecture. Here, our objective is to achieve comparable accuracy by using a pure CNN architecture without ensemble architecture, as ensemble architectures introduce increased computational cost and high testing complexity. Thus, a CNN architecture is proposed in order to achieve accuracy even better than that of ensemble architectures, along with reduced operational complexity and cost. Moreover, we also present an appropriate combination of learning parameters in designing a CNN that leads us to reach a new absolute record in classifying MNIST handwritten digits. We carried out extensive experiments and achieved a recognition accuracy of 99.87% for a MNIST dataset.

This study addresses the problem of the automatic detection of disease states of the retina. In order to solve the abovementioned problem, this study develops an artificially intelligent model. The model is based on a customized 19-layer deep convolutional neural network called VGG-19 architecture. The model (VGG-19 architecture) is empowered by transfer learning. The model is designed so that it can learn from a large set of images taken with optical coherence tomography (OCT) and classify them into four conditions of the retina: (1) choroidal neovascularization, (2) drusen, (3) diabetic macular edema, and (4) normal form. The training datasets (taken from publicly available sources) consist of 84,568 instances of OCT retinal images. The datasets exhibit all four classes of retinal disease mentioned above. The proposed model achieved a 99.17% classification accuracy with 0.995 specificities and 0.99 sensitivity, making it better than the existing models. In addition, the proper statistical evaluation is done on the predictions using such performance measures as (1) area under the receiver operating characteristic curve, (2) Cohen’s kappa parameter, and (3) confusion matrix. Experimental results show that the proposed VGG-19 architecture coupled with transfer learning is an effective technique for automatically detecting the disease state of a retina.

Character Segmentation is the most crucial step for any OCR (Optical Character Recognition) System. The selection of segmentation algorithm being used is the key factor in deciding the accuracy of OCR system. If there is a good segmentation of characters, the recognition accuracy will also be high. Segmentation of words into characters becomes very difficult due to the cursive and unconstrained nature of the handwritten script. This paper proposes a new vertical segmentation algorithm in which the segmentation points are located after thinning the word image to get the stroke width of a single pixel. The knowledge of shape and geometry of English characters is used in the segmentation process to detect ligatures. The proposed segmentation approach is tested on a local benchmark database and high segmentation accuracy is found to be achieved.

Characters extraction is the most critical pre-processing step for any off-line text recognition system because the characters are the smallest unit of any language script. The paper proposes an approach to segment character images from the text containing images and computer printed or handwritten words. This segmentation app roach is based on a set of properties for each connected component (object) in the whole binary image of the machine printed or handwritten text containing some other images. These words which are printed along with some images are of different lengths and are printed by different cursive fonts of different sizes. This character extraction technique is applied for the segmentation of untouched characters from the machine printed or handwritten words of varying length written on a noisy background having some images etc. Very promising results are achieved which reveals the robustness of the proposed character detection and extraction technique.