Kanthali Harshada, Mr. A. A. Dalvi, Prof. J. R. Mahajan · International Journal of Advanced Research in Science Communication and Technology 2026 · 2026
DOI: 10.48175/ijarsct-38889
Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).
Computer vision enables computer systems to extract useful meaning from images and video. Recent advances in artificial intelligence, particularly deep learning, have improved the ability of machines to recognise objects, interpret scenes, and identify visual patterns. This paper presents a structured review of AI techniques used in computer vision, with emphasis on convolutional neural networks (CNNs), their main processing stages, practical applications, and continuing limitations. It explains why manually designed visual features can be unreliable when images vary in lighting, scale, viewpoint, or background, and how learned representations help address these problems. Applications in healthcare, transportation, astronomy, and industrial inspection are discussed alongside concerns involving data quality, bias, privacy, explainability, and computational cost. The paper is a conceptual review based on established literature and does not claim a new experiment or measured performance result. It concludes that responsible deployment requires not only accurate models, but also representative data, careful evaluation, human oversight, and monitoring after deployment
No comments yet — start the discussion below.