In order to solve the problem of low degree of mechanical automation, such as manual insertion and arrangement of industrial cable automatic winding system, a new type of automatic cable winding system based on machine vision is innovatively established. The system processes video image sets collected by industrial camera to obtain target coordinates, and the PC controls the rotation of the mechanical arm to complete the process of plugging and arranging, thereby improving the degree of automation. A target recognition algorithm based on Hu invariant moment and multi-degree-of-freedom neural network (MDOFNN) is designed when searching for targets. The algorithm recognizes the target contour information and obtains the target coordinates by ellipse fitting method. The target recognition algorithm and the support vector machine (SVM) are compared by experiments. The results show that the recognition rate of this algorithm is higher than that of SVM, which is suitable for industrial inspection.
[1]
张兰芳,et al.
Face recognition method based on neural network
,
2012
.
[2]
Jan Urban,et al.
Application of machine vision systems in aquaculture with emphasis on fish: state-of-the-art and key issues
,
2017
.
[3]
Md. Nurul Kabir,et al.
A Survey on Image Segmentation Methods
,
2016
.
[4]
Maria Cristina Timar,et al.
Image Processing Method as a Supporting Tool for Wood Species Identification
,
2013
.
[5]
Lai Wei,et al.
Constructive Algorithm of Multi-Weight Neuron Covering Networks
,
2012
.