Downloads: 1 | Views: 370 | Weekly Hits: ⮙1 | Monthly Hits: ⮙1
New Innovation and Idea | Electrical & Electronics Engineering | India | Volume 11 Issue 11, November 2022 | Rating: 4.6 / 10
Object Sorting System using Visual Inspection through Computer Vision
Daksha Sunil Saraf
Abstract: Object tracking is one of the major fundamental challenging problems in computer vision applications as difficulties in tracking objects can arise due to intrinsic and extrinsic factors like deformation, camera motion, motion blur and occlusion, this project presents a helpful application with a real-time object detection system that can automatically capture the user-defined important objects and segregate them instantly with the help of a rack and pinion mechanism. The problem discussed in this article will be solved using object detection using deep neural network especially convolutional neural networks. We would also try to present an embedded technique for real-time object detection and recognition that runs at high frames per second (FPS) by applying deep learning for computer vision. To address the real-life problems where, instant accurate decision taking capability of a robot is not yet achieved. For this, object detection is the new lookout. A camera module used to track objects. But object tracking is one of the major fundamental challenging problems in computer vision applications. There are many difficulties in tracking objects which arise due to intrinsic and extrinsic factors like deformation, camera motion, motion blur and occlusion. This project presents a helpful application with a real-time object detection system that can automatically capture the user-defined important objects. By identifying these users defined objects we shall segregate them instantly with the help of a robotic arm. The problem discussed in this article will be solved using object detection using deep neural network especially convolutional neural networks. We would also try to present an embedded technique for real-time object detection and recognition that runs at high frames per second (FPS) by applying deep learning for computer vision. Thus, by integrating computer vision applications with deep learning and robotics, we can achieve a more reliable, accurate and time conserving technology for the industries with heavy machinery.
Keywords: Object Tracking, Hardware System, Conveyor
Edition: Volume 11 Issue 11, November 2022,
Pages: 192 - 196