Font Size: a A A

Development Of Contour Detection And Cutting Track Optimization Algorithm For Large-face Leather

Posted on:2022-01-08Degree:MasterType:Thesis
Country:ChinaCandidate:Y LiuFull Text:PDF
GTID:2481306539959359Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
In recent years,with the improvement of tanning technology,the quality of two-layer leather products has been close to that of first-layer leather products.In order to enhance the characteristics of the two layers of skin,a layer of PU film is usually applied to the surface of the skin.At present,domestic small and medium-sized enterprises are still mainly using manual cutting and redundant film covering,which has the problem of low processing efficiency and raw material utilization rate.When the cutting machine is processed automatically,if it is cut in accordance with the trajectory detected by the original contour of the two layers of skin,it will be difficult to cut because of the existence of a large number of small line segments in the processing trajectory,and it is necessary to frequently lift the knife and drop the knife when it encounters a sharp corner in the cutting process,which is inefficient.In view of these problems,the design of contour detection and cutting track optimization algorithm for leather was carried out in this paper.The main research contents are as follows:(1)A complete two-layer leather lamination cutting process was designed.According to the characteristics of the two-layer leather lamination cutting process,the hardware selection of industrial computer,camera,lens,light source and bracket was carried out,and the corresponding debugging work was carried out to build a set of visual inspection equipment with leather cutting machine tools.(2)The conversion process between coordinate systems is introduced,that is,hand-eye calibration,and the Canny operator is improved.For image denoising,an improved bilateral filter is proposed to replace the Gaussian filter,which can deal with pepper and salt noise and Gaussian noise at the same time.The high and low threshold of the gradient can be obtained adaptively by using the idea of Otsu method.Then refine the contour obtained by using the improved Canny operator and trace the chain code.(3)In order to get the machining track of fast cutting,the improved piecewise least square arc straight line fitting method was used to fit the point set of straight line arc,and compared with the traditional fitting method,the number of fitting line segments was reduced,the running time of the algorithm was reduced,and the beginning and end line segments could be connected.Aiming at the problem of frequently lifting and dropping knives,an adaptive corner transition algorithm is proposed to improve the stability and efficiency of machining.(4)The software of the leather contour detection and trajectory optimization system was developed.The overall architecture of the system software was designed.According to the actual requirements of leather detection and processing,the functional modules of the system software were divided and developed.(5)The simulation experiment of the two-layer skin processing was designed,and the simulation results showed that the average speed of the whole processing was improved and the processing time was reduced.Then,the actual processing experiment of two layers of leather was carried out on a leather cutting machine.The experimental results showed that the algorithm of this topic improved the efficiency of film laminating cutting,improved the utilization rate of leather,and optimized the production process.
Keywords/Search Tags:Double skin, Image denoising, Edge detection, Straight-line arc fitting, The corner transition
PDF Full Text Request
Related items