Journal of Systems Engineering and Electronics ›› 2018, Vol. 29 ›› Issue (6): 1300-1307.doi: 10.21629/JSEE.2018.06.17

• Software Algorithm and Simulation • Previous Articles     Next Articles

Real-time object segmentation based on convolutional neural network with saliency optimization for picking

Jinbo CHEN(), Zhiheng WANG(), Hengyu LI*()   

  • Received:2017-09-22 Online:2018-12-25 Published:2018-12-26
  • Contact: Hengyu LI E-mail:jbchen@shu.edu.cn;hengzz@i.shu.edu.cn;lihengyu@shu.edu.cn
  • About author:CHEN Jinbo was born in 1980. He received his M.S. and Ph.D. degrees, both in mechatronic engineering from Shanghai University, Shanghai, China, in 2005 and 2014 respectively. He is currently working as a lecturer in School of Mechatronic Engineering and Automation at Shanghai University. His research interests include computer vision and machine learning. E-mail: jbchen@shu.edu.cn|WANG Zhiheng was born in 1993. In 2015, he enrolled in Shanghai University with a major of mechatronic engineering. He is pursuing his academic master's degree in Shanghai University. He has experienced in some projects about control and vision inspection system. Especially in the machine vision, he has deep research experience. His research interests are in machine vision theory and technology. E-mail: hengzz@i.shu.edu.cn|LI Hengyu was born in 1983. He received his B.S. degree in mechanical engineering and automation from Henan Polytechnic University, China, in 2006, and M.S. and Ph.D. degrees in mechanical and electronic engineering from Shanghai University, China, in 2009 and 2012, respectively. He is currently an associate professor with the School of Mechatronic Engineering and Automation, Shanghai University. His research interests include mechatronics and robot bionic vision system autonomous cooperative control for multiple robots. E-mail: lihengyu@shu.edu.cn
  • Supported by:
    the National Natural Science Foundation of China(61233010);the National Natural Science Foundation of China(61305106);the Shanghai Natural Science Foundation(17ZR1409700);the Shanghai Natural Science Foundation(18ZR1415300);the basic research project of Shanghai Municipal Science and Technology Commission(16JC1400900);This work was supported by the National Natural Science Foundation of China (61233010; 61305106), the Shanghai Natural Science Foundation (17ZR1409700; 18ZR1415300) and the basic research project of Shanghai Municipal Science and Technology Commission (16JC1400900)

Abstract:

This paper concerns the problem of object segmentation in real-time for picking system. A region proposal method inspired by human glance based on the convolutional neural network is proposed to select promising regions, allowing more processing is reserved only for these regions. The speed of object segmentation is significantly improved by the region proposal method. By the combination of the region proposal method based on the convolutional neural network and superpixel method, the category and location information can be used to segment objects and image redundancy is significantly reduced. The processing time is reduced considerably by this to achieve the real time. Experiments show that the proposed method can segment the interested target object in real time on an ordinary laptop.

Key words: convolutional neural network, object detection, object segmentation, superpixel, saliency optimization