Search

Recognizing specific characters from a large set of handwritten characters is called off-line character recognition. Devanagari, the reading and writing script, is widely used in a broad region of India. This paper proposes using the modified LeNet-5 architecture for recognizing Devanagari off-line handwritten characters using the Devanagari Handwritten Character Dataset (DHCD). The dataset comprises 46 distinct character classes, each with 2000 images. The proposed modified Lenet-5 Base Architecture (MLCNN8) uses more filters and kernels of small sizes with batch normalization. The proposed modified Lenet-5 Base Architecture (MLCNN8) outperformed the Lenet-5 Base Architecture (LCNN) with an accuracy of 99.21% in 53 epochs, whereas LCNN achieved an accuracy of 94.50% in 48 epochs. Hence, the proposed modified Lenet-5 architecture outperforms the Lenet-5 Base Architecture. The proposed model's results were compared with other state-of-the-art and found that it also has an edge for recognizing the Devanagari handwritten characters. [ABSTRACT FROM AUTHOR]

Mushrooms are fungi that play significant roles in ecosystems, food, and medicine. However, distinguishing toxic mushrooms from non-toxic ones is challenging, as their visual characteristics can be misleading. Deep learning, a subset of artificial intelligence, offers a promising approach to detecting and classifying mushrooms based on their toxicity. By training deep neural networks on datasets of labelled mushroom images, it becomes possible to create models capable of accurately differentiating between toxic and non-toxic mushrooms. Mushrooms are fungi that play significant roles in ecosystems, food, and medicine. However, manually distinguishing toxic mushrooms from non-toxic ones is a challenging and time-consuming task. This research work is to help the local farmers and exporters to identify the difference between the toxic and the edible mushroom which will eventually boost production and will directly have an impact on the economic growth in India. This paper compared different deep learning algorithms such as Support Vector Machine (SVM), Resnet50, YOLO V5 and Alexnet, which have been found to be the most useful in classifying the mushrooms effectively as toxic and non-toxic respectively. The patterns in the images of the mushrooms have been predicted through the above classifiers. The validation techniques show SVM to be the most effective with an accuracy of 83%. [ABSTRACT FROM AUTHOR]

Environment, price, regulation, and other factors influence the price of agricultural products, which is a social signal of product supply and demand. The price of many agricultural products fluctuates greatly due to the asymmetry between production and marketing details. Horticultural goods are particularly price sensitive because they cannot be stored for long periods of time. It is very important and helpful to forecast the price of horticultural products which is crucial in designing a cropping plan. The proposed method guides the farmers in agricultural product production and harvesting plans. Farmers can benefit from long-term forecasting since it helps them plan their planting and harvesting schedules. Customers can also profit from daily average price estimates for the short term. This paper study the time series models such as ARIMA, SARIMA, and neural network models such as BPN, LSTM and are used for wheat cost prediction in India. A large scale available data set is collected and tested. The results shows that since ARIMA and SARIMA models are well suited for smallscale, continuous, and periodic data, the BPN and LSTM provide more accurate and faster results for predicting well weekly and monthly trends of price fluctuation. [ABSTRACT FROM AUTHOR]

Tamizhi (Tamil-Brahmi) script is one of the oldest scripts in India from which most of the modern Indian scripts are evolved. The ancient historical documents are generally preserved as digitised texts using Optical Character Recognition (OCR) technique. But the development of OCR for Tamizhi documents is highly challenging as many characters have similar shapes and structures with very small variations. In specific, for Tamizhi script it is very difficult to build an OCR as many characters are combined characters. This can be a single character formed by a single vowel/consonant or compound characters formed by combining vowels and consonants. This paper deals with the development of Tamizhi OCR for printed Tamizhi documents which is anticipated to perform efficiently irrespective of poor quality, noises and various input formats of Tamizhi documents. This is a preliminary study towards developing an OCR for handwritten Tamizhi inscription images that recognises text captured from onsite inscriptions. The developed Tamizhi OCR for printed text can produce an accuracy of about 91.12 per cent. [ABSTRACT FROM AUTHOR]

In this paper, we propose a decision support system for Indian rice farmers for identifying diseases. In a country like India, food security is an essential concern. Additionally, diseases in plants can cause a significant loss. Early-stage detection of diseases can help in improving the production of rice. In this context, first we investigate the recent contributed efforts in the field of plant disease detection by analysing plant leaves using machine learning and image processing techniques. Next, the datasets and relevant algorithms are concluded. Then, a machine learning model has been presented. The model includes the edge feature extraction using canny edge detection technique, colour features are extracted using grid colour movement, and the texture analysis is performed using Local Binary Pattern (LBP). In the next step, using the extracted features, we have prepared a combined feature vector to train the Machine Learning (ML) algorithms namely Support Vector Machine (SVM) and Artificial Neural Network (ANN). These machine learning algorithms are organized in such a manner that the proposed decision support model can identify and differentiate the leaf plants. Additionally, it also recognizes the rice plants when we query. Secondly, the model is also able to recognize rice plant diseases. The first scenario of the experiment has been carried out using Plant Village dataset. The second scenario of experiment uses the rice plant disease dataset obtained from Kaggle with three classes. The second dataset used which is known as the Mendeley dataset which contains five different diseases as class labels. The experimental study with the implemented system confirms the superiority of ANN to be used with the proposed decision support system as compared to the SVM algorithm in terms of accuracy and time consumption. Finally, future work has also been highlighted. [ABSTRACT FROM AUTHOR]

Morphological processing of Indian languages is one of the most escalating fields in the era of Natural Language Processing (NLP) since the last decade. The evaluation of Asian languages is a highly relevant field in the times of text mining and information retrieval. The morphological evaluation of a text can be employed for extraction and classification of knowledge. This paper amalgamates morphological evaluation and sentiment prediction of Punjabi language text. The textual data for Punjabi language is concerned with farmer suicide cases reported for Punjab state of India. The pre-processing phase of this study involves morphological evaluation and normalization of Punjabi words to their respective canonical forms. The next phase carries out training and testing of deep neural network model on refined Punjabi tokens obtained from the earlier phase. The proposed model classifies Punjabi tokens into four negatively oriented classes tailored for farmer suicide cases. The average accuracies of sentiment prediction obtained after 10-fold cross validation are 93.85%, 88.53%, 83.3%, and 95.45% for the four respective classes. The proposed framework yields satisfactory results on 275 Punjabi text documents with the overall accuracy of 90.29% for sentiment classification. [ABSTRACT FROM AUTHOR]

The novel coronavirus (COVID-19), which is one of its kind of humanitarian disasters, has affected people and businesses worldwide, triggering a global economic crisis. In this aspect, the tourism sector is not being left behind. The pandemic has not only affected the foreign exchange earnings (FEE) but also affected various regional developments, job opportunities, thereby disrupting the local communities as a whole. As there has been a substantial decline in the arrivals of overseas tourists in India in 2020, the paper aims to predict foreign tourists' arrival in India and FEE using artificial neural networks (ANN). Furthermore, we analyse the impact of COVID-19 based on four scenarios considering with and without lockdown in terms of loss and gain in FEE. Lastly, the results obtained will help policymakers make necessary strategic and operational decisions, along with maximizing the FEE. [ABSTRACT FROM AUTHOR]

In this paper, deep learning is employed to propose an Artificial Neural Network (ANN) based online incremental learning technique for developing an adaptive and non-intrusive analytical model of Covid-19 pandemic to analyze the temporal dynamics of the disease spread. The model is able to intelligently adapt to new ground realities in real-time eliminating the need to retrain the model from scratch every time a new data set is received from the continuously evolving training data. The model is validated with the historical data and a forecast of the disease spread for 30-days is given in the five worst affected states of India. [ABSTRACT FROM AUTHOR]

Corn is one of the most popular food grains in the India and crop loss due to diseases substantially affects the Indian economy and threatens the food availability. Recent access of smart devices can be utilized to provide automatic diagnosis of corn diseases and prevent severe crop losses. This paper presents a real time method based on deep convolutional neural network for corn leaf disease recognition. Deep neural network performance is improved by tuning the hyper-parameters and adjusting the pooling combinations on a system with GPU. Further, the number of parameters of the developed model is optimized to make it suitable for real time inference. The pre-trained deep CNN model was deployed onto raspberry pi 3 using Intel Movidius Neural Compute Stick consisting dedicated CNN hardware blocks. During the recognition of corn leaf diseases, the deep learning model achieves an accuracy of 88.46% demonstrating the feasibility of this method. The presented corn plant disease recognition model is capable of running on standalone smart devices like raspberry-pi or smart-phone and drones. [ABSTRACT FROM AUTHOR]

Abstract The exact measure of mitotic count is one of the crucial parameters in breast cancer grading and prognosis. Detection of mitosis in standard H & E stained histopathology images is challenging due to diffused intensities along object boundaries and shape variation in different stages of mitosis. This paper explores the feasibility of transfer learning for mitosis detection. A pre-trained Convolutional Neural Network is transformed by coupling random forest classifier with the initial fully connected layers to extract discriminant features from nuclei patches and to precisely prognosticate the class label of cell nuclei. The modified Convolutional Neural Network accurately classify the detected cell nuclei with limited training data. The designed framework accomplishes higher classification accuracy by carefully fine tuning the pre-trained model and pre-processing the extracted features. Moreover, proposed method is evaluated on MITOS dataset provided for the MITOS-ATYPIA contest 2014 and clinical data set from Regional Cancer Centre, Thiruvananthapuram, India. Significance of Convolutional Neural Network based method is justified by comparing with recently reported works including a Multi Classifier System based on Deep Belief Network. Experiments show that the pre-trained Convolutional Neural Network model outperforms conventionally used detection systems and provides at least 15% improvement in F-score on other state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

Indian University Admission is a complex phenomenon encompassing various factors both tangible and intangible. Apart from Merit - Caste, Community and Religion play a crucial role in getting admission to various courses offered by Universities. The Single Window Admission System followed by almost all Government Universities (for example: Anna University, Chennai) is, so far, the best practice to offer Admission sought by student community. But, still, in Private Universities, the Admission process is done manually since number of students seeking a course is smaller in size. This manual process is fraught with some drawbacks: such as slower in time, cumbersome and costlier, bias by the Admission Officer, manual errors while processing, due to influence exercised by powerful people qualified students not getting their course of choice and unqualified students getting into their course of willingness and etc. This paper addresses these problems via neural network architecture based Admission system which will eliminate all the pitfalls and drawbacks inherent in the current system and offers a smooth, clearer, easier and cost effective way of student admission system implementation in Indian Universities. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents the methodologies for incipient fault detection in power transformers both off-line and on-line. An artificial neural network is used to detect faults off-line with dissolved gas analysis reports of transformers and whereas wavelet transforms are being used for on-line fault detection. The accuracy in fault detection through artificial neural networks is compared with Rogers ratio method using the analysis of experimental oil samples for power transformers of power companies in Andhra Pradesh, India. The Wavelet transform techniques have been developed with different mother wavelets to detect incipient faults and to distinguish between incipient fault and short circuit fault. Further their performances with different mother wavelets are compared. [Copyright &y& Elsevier]

Abstract: Seed planting equipment with inclined plate seed metering devices is the most commonly used equipment for planting of peanut crop in India. For obtaining the high yield, it is very essential to drop the peanut seeds in rows maintaining accurate seed rate and seed spacing with minimum damage to seeds during metering. This mainly depends on the forward speed of the planting equipment, rotary speed of the metering plate and the area of cells on the plate. The relationship between these factors and the performance parameters viz., seed rate, seed spacing and percent seed damage can be established using regression analysis. But they may not be very accurate and consistent throughout the solution space. Hence, an attempt has been made in this paper to develop the feed forward artificial neural network (ANN) models for the prediction of the performance parameters of an inclined plate seed metering device. The data were generated in the laboratory by conducting experiments on a sticky belt test stand provided with a seed metering device and an opto-electronic seed counter. The generated data was used to develop both statistical and neural network models. The optimum architecture of the neural network models was determined using genetic algorithm (GA) as a single objective constrained optimization problem. The performance of the developed models was compared among themselves for 4 randomly generated test cases. The results show that the ANN model predicted the performance parameters of the seed metering device better than the statistical models. It is possible to determine the optimum levels of the input parameters to obtain the desired performance parameters of the seed metering device by performing reverse mapping of the developed ANN models. [Copyright &y& Elsevier]