Scholarlink

<p>Object recognition is crucial in computer vision, enabling applications like autonomous vehicle navigation, inventory management, medical image processing, robotics, traffic signal verification, and virtual reality. Yet, factors such as lighting conditions and perspectives can impact the quality of object detection in both standalone images and real-time feeds. This paper reviews recent object recognition models, concentrating on challenges related to lighting and perspectives. It offers an in-depth analysis of contemporary deep learning models for object detection, highlighting their architectures and performance. The study also scrutinizes the different datasets used in these models. Finally, it identifies research gaps and proposes potential solutions for future research directions in computer vision to enhance object recognition efficiency.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhancing object recognition: a comprehensive analysis of CNN based deep learning models considering lighting conditions and perspectives

  • 期刊论文

  • 发表日期：

    2025年6月21日

查看全文
  • Penumala Nani,
  • Smita Das,
  • Sayeli Dey

摘要

Object recognition is crucial in computer vision, enabling applications like autonomous vehicle navigation, inventory management, medical image processing, robotics, traffic signal verification, and virtual reality. Yet, factors such as lighting conditions and perspectives can impact the quality of object detection in both standalone images and real-time feeds. This paper reviews recent object recognition models, concentrating on challenges related to lighting and perspectives. It offers an in-depth analysis of contemporary deep learning models for object detection, highlighting their architectures and performance. The study also scrutinizes the different datasets used in these models. Finally, it identifies research gaps and proposes potential solutions for future research directions in computer vision to enhance object recognition efficiency.