Research On The Technology Of Machine Evaluation And Classification Based On The Assembly Quality Of Standard Chassis

Chassis standard parts are the key protection structure of electronic equipment,data communication equipment and information technology equipment.The assembly of the chassis has characteristics such as multi-unit,complexity of the standard parts and difference of the structure,which challenge the detection of assembly quality.Aiming at the methods to extract image features of chassis standard parts assembly quality,this paper,"Image Feature Extraction and Construction Method for Assembly Quality of Chassis Standard Parts",studies the image feature extraction method of standard parts chassis based on multi corners combination,image feature extraction method of semi supervised standard parts chassis based on PI-SURF,sample library and form template method of standard chassis image.The research and development of a universal chassis assembly quality inspection template,which has the important academic value and practical significance,realize the automation of the standard parts chassis detection and promote the development of intelligent manufacturing technology.The research work was supported by Guangzhou Science and Technology Program(No.201701016064,201509010008).Image feature extraction method of standard parts chassis based on multi corners combination,image feature extraction method of semi supervised standard parts chassis based on RPI-SURF,sample library and form template method of standard chassis image are proposed.The validity and advancement of image feature extraction and construction method of standard parts chassis are verified by experiments.Main research work of this paper includes followings:(1)Based on the analysis of image characteristics and corner features of three kinds of chassis standard parts,it is pointed out that it is suitable for the recognition and classification of image corner features.It analyzes the mechanism of Harris,Shi-Tomasi,FAST algorithm of corner extraction,obtains the corner feature on the chassis standard parts and studies on the distribution and characteristics of corner feature of chassis standard parts.A method of image feature extraction based on multi corners combination is proposed.(2)According to the feature of the space location of the assembly parts of the chassis,the method of dividing the detection area is designed.Combined with the SURF feature extraction method,this paper studies the region division technology of the chassis standard parts assembly quality detection region based on PI-SURF,reducing the computation of image detection and improving the detection accuracy.(3)Based on the analysis of the information of the chassis detection template,the relationship between the image sample library of the chassis and the chassis quality inspection template are analyzed,the principle and method of the chassis assembly quality inspection template are discussed,the data format of template id studied and the construction process of the multi type image sample library and the standard image template are described with the corresponding case of assembly quality inspection template.(4)To build the chassis standard parts assembly platform,this paper uses ADLink MXC-2601 D industrial embedded chassis for image feature extraction and template construction test,extracting the image corner feature and building the chassis assembly quality inspection template.