This document analyses the surface of piston in two aspects, discrete points of axial section profile are fitted by cubic spline tool of MATLAB, interpolate curve after fitting with method of precision linear approximation. Equipartition of cross section profile are done,and through subdivision of points of equipartition,to approach theory contour. the number of divide points must satisfy the request was made, with the principle of the relative movement，in the processing of cross section in an ellipse，detailed analysis the motion of line motor，pointed out how changes the speed and acceleration of workbench.and must satisfy the basic conditions of the line motor to respond to the turning of cross section in an ellipse. On this basis, design a lathe machining CNC system of piston with structure of upper and lower computer, moulding of surface of piston are completed by upper computer, feed drive of X, Z axis and spindle are completed by lower computer, surface of middle-convex and varying piston are processed finally.

This dissertation mainly introduces Fuzzy-PID Control which are used in the PFPL (Pseudo-flexible Production Line). The servo system plays an important role in the NC machine. The performance of servo system decide the quality of NC machine in very great degree This dissertation research and develop a milling and paring manufacture unit of open CNC system -- Numerical control unit of PFPL. By using the GT-SV motion controller, put up the software and hardware platform of NC unit rapidly with module design method and the target -oriented programming idea .Take NC milling machine as the process platform, this dissertation use intellectual PID control method to improve the servo control of motion controller, through the pointing accuracy survey of the milling machine, confirmed the actual effect of Fuzzy-PID control.

The paper analyses the problem of beer bottles detection techniques on the beer bottles production line, uses digital image processing technique on the beer bottles online defect detection. The paper puts forward the designing ideas of the hardware, developing flow of the software and the algorithm of beer bottles detection. TMSDM642 is used to set up the real-time video processing system of the hardware .The hardware system is mainly composed of three parts: the part of memory, the part of the input and the part of the output. When beer bottles are put into the work area, the video images of the bottle-mouth and bottle-bottom will be gained by the CCD camera, firstly, preprocessing is used to eliminate video image noise. Secondly, the image segmentation algorithm is used to detect defects in video images. Lastly the goal of extracting defects will be accomplished. The experimental result indicated that this system may effectively exam the flaw or the unqualified beer bottles.

The current deficiencies in domestic vaccine embryo is mainly dependent on artificial visual identification to identify, This has resulted in the detection speed and accuracy is not high, This has resulted in the detection speed and accuracy is not high, poor security features. In order to change the traditional manual operation mode, this article put forward the concept of image detection platform. Hardware aspect through CCD camera gain image, Takes the imagery processing by the core of TMS320DM642, the software aspect introduces operating system's programming thought to the imagery processing each function modulation, the system imagery processing uses VC++ to confirm the examination algorithm in advance, to reduce the development cycle. The experimental result indicated that this system changed the traditional manual control pattern, raised the examination speed and the rate of accuracy.

This paper mainly studies a decision tree method based on support vector machine to identify egg embryo. It analyses the main characteristics of the types of egg embryo and sets up a kind of multilayer decision tree classifier by using "solution space". And it figures out the correct rate of the decision tree classifier, which concentrates on the types of egg embryo. By introducing the support vector machine (SVM) algorithm based on the structure of the binary tree for multi-class classification, it identifies different kinds of egg embryo. Not only does this method make the decision capability of the optimal one in every level of the decision tree, but also assures the overall optimal performance of the whole decision tree, and effectively improves the correct recognition rate of the decision tree classifier about the types of egg embryo.