Manzoor Ahmed Hashmani

Papers

Total Citations

H-Index

About

Manzoor Ahmed Hashmani is a leading researcher in computer vision and educational data mining, whose work bridges the gap between algorithmic precision and real-world applicability. His most impactful contribution comes from the domain of semantic segmentation, where he developed the Unified DeepLabV3+ architecture—a pioneering framework designed to tackle the notoriously difficult problem of semi-dark image semantic segmentation. This work, which has garnered 15 citations, addresses a critical bottleneck in autonomous visual perception by enabling deep convolutional neural networks, specifically ResNet-based models, to classify dynamic scenes with high accuracy under low-light conditions. In parallel, Hashmani has made significant strides in educational data mining (EDM), authoring a comprehensive review on feature selection methods that improve classification performance in student academic prediction. This research, cited 4 times, provides educational managers with actionable insights to identify key factors influencing student success. By combining rigorous deep learning innovation with practical data-driven solutions for education, Hashmani’s work demonstrates a rare versatility—advancing both the theoretical foundations of computer vision and the applied analytics that shape modern learning environments.