Scholarlink

In computer vision, identifying objects is a challenging task with applications in robotics, monitoring, and automated vehicles. Nonetheless, recognizing objects tasks require greater analysis skill, energy consumption, and computational power than image categorization. Our project uses a MATLAB-implemented Convolutional Neural Network to create a real-time object identification module. The suggested system will make use of CNNs, a type of deep learning technology, to precisely locate and identify objects in live video streams. This research proposes a hybrid-YOLOv4 recognition algorithm-based approach for object classification and detection. Convolutional neural networks are able to extract features from photos and track objects with remarkable accuracy. DarkNet is employed as the backbone to enhance feature interchange and reusing, while strategies like Bag-of-Specials and Bag-of-Freebies are implemented for item identification. As a result, the enhanced network architecture optimizes speed and recognition accuracy.

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

Assessment of Real-Time Object Identification Using Convolutional Neural Networks with YOLO V4

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Y. Sailaja,
  • T. Lavanya,
  • B. Kavitha Rani,
  • Sanjana S. Nazare,
  • M. D. Rafeeq,
  • Maredu Mallikarjun

摘要

In computer vision, identifying objects is a challenging task with applications in robotics, monitoring, and automated vehicles. Nonetheless, recognizing objects tasks require greater analysis skill, energy consumption, and computational power than image categorization. Our project uses a MATLAB-implemented Convolutional Neural Network to create a real-time object identification module. The suggested system will make use of CNNs, a type of deep learning technology, to precisely locate and identify objects in live video streams. This research proposes a hybrid-YOLOv4 recognition algorithm-based approach for object classification and detection. Convolutional neural networks are able to extract features from photos and track objects with remarkable accuracy. DarkNet is employed as the backbone to enhance feature interchange and reusing, while strategies like Bag-of-Specials and Bag-of-Freebies are implemented for item identification. As a result, the enhanced network architecture optimizes speed and recognition accuracy.