Open Access

Article

Article ID: 545

Plant leaf disease classification using FractalNet

by Hmidi Alaeddine, Malek Jihene

Computing and Artificial Intelligence, Vol.2, No.2, 2024; 36 Views, 16 PDF Downloads

In this work, an effort is made to apply the FractalNet model in the field of plant disease classification. The proposed model was trained and tested using a “PlantVillage” plant disease image dataset using a central processing unit (CPU) environment for 300 epochs. It produced an average classification accuracy of 99.9632% on the test dataset. The experimental results demonstrate the efficiency of the proposed model and show that the model achieved the highest values compared to other deep learning models in the PlantVillage datasets.

Open Access

Article

Article ID: 1427

The computational analysis of COVID-19-induced socio-economic, environmental, and climatic disruptions on the Indian food production system

by Adya Aiswarya Dash, Sonu Sharma

Computing and Artificial Intelligence, Vol.2, No.2, 2024; 32 Views, 7 PDF Downloads

COVID-19 dominantly affected all the sectors of the Indian economy, surprisingly the impact is much lower with respect to the agricultural production (−2.7%) in India. The increase in yield of the crops can be attributed to the variables such as environmental, climatic, and socio-demographic factors. The study illustrates the relationship of the increase in crop yield in India during the first wave of COVID-19 along with the increase in the infection count and the land under cultivation attributed to supporting factors during the year 2020. The relation is explained by the method of ordinary least square (OLS) and geographically weighted regression (GWR). The distribution of the increase in crop yield across India is analyzed against COVID-19 infections along with other dominant factors. Useful intuitions against crop yield can be generated by studying the spatial relationships between them. The geographically weighted regression method depicted an increase in R 2 value as compared to the global ordinary least regression method. The Akaike information criterion in the geographically weighted regression method is also lower as compared to the ordinary least square therefore explaining GWR as a better model as compared to OLS. The combination of the various variables affecting agricultural yield in India is found to vary geographically as well as with the type of crops.

Open Access

Article

Article ID: 1409

Predicting manipulated regions in deepfake videos using convolutional vision transformers

by Mohan Bhandari, Sushant Shrestha, Utsab Karki, Santosh Adhikari, Rajan Gaihre

Computing and Artificial Intelligence, Vol.2, No.2, 2024; 195 Views, 34 PDF Downloads

Deepfake technology, which uses artificial intelligence to create and manipulate realistic synthetic media, poses a serious threat to the trustworthiness and integrity of digital content. Deepfakes can be used to generate, swap, or modify faces in videos, altering the appearance, identity, or expression of individuals. This study presents an approach for deepfake detection, based on a convolutional vision transformer (CViT), a hybrid model that combines convolutional neural networks (CNNs) and vision transformers (ViTs). The proposed study uses a 20-layer CNN to extract learnable features from face images, and a ViT to classify them into real or fake categories. The study also employs MTCNN, a multi-task cascaded network, to detect and align faces in videos, improving the accuracy and efficiency of the face extraction process. The method is assessed using the FaceForensics++ dataset, which comprises 15,800 images sourced from 1600 videos. With an 80:10:10 split ratio, the experimental results show that the proposed method achieves an accuracy of 92.5% and an AUC of 0.91. We use Gradient-Weighted Class Activation Mapping (Grad-CAM) visualization that highlights distinctive image regions used for making a decision. The proposed method demonstrates a high capability of detecting and distinguishing between genuine and manipulated videos, contributing to the enhancement of media authenticity and security.

Open Access

Article

Article ID: 1364

Software cost estimation tool: A App based application, estimate the cost of software project

by Ajay Jaiswal, Piyush Malviya, Lucky Parihar, Rani Pathak, Kuldeep Rajput

Computing and Artificial Intelligence, Vol.2, No.2, 2024; 22 Views, 1 PDF Downloads

This paper presents the design and implementation of a software cost estimation tool integrated into a mobile application developed using Flutter. The tool incorporates various techniques for software cost estimation, including expert judgment, function point analysis, 3D point analysis, and the COCOMO model. The purpose of the program is to give software engineers and project managers a practical and effective tool for calculating the time and money needed for software development projects. The paper provides a thorough explanation of each estimation technique’s implementation, along with a discussion of the app’s main features and functionalities. Because of the app’s intuitive and user-friendly design, users can quickly enter project data and get precise cost estimates. The tool’s efficacy is assessed using case studies and contrasts with other software cost estimation methods currently in use. The outcomes show that the app can produce trustworthy and precise cost estimates, which makes it an important resource for software development projects.