Dissertation > Excellent graduate degree dissertation topics show

Outline Detection and Brand Recognition of Link-plate Based on Image Processing

Author: XieHuaLong
Tutor: TangKeHong
School: Jilin University
Course: Mechanical and Electronic Engineering
Keywords: Link-plate median filter edge detection state machine outline FPGA ARM
CLC: TP391.41
Type: Master's thesis
Year: 2006
Downloads: 185
Quote: 1
Read: Download Dissertation

Abstract


Along with the enhancement of the industrial production level, today,the society demands more product as well as better product quality, and theautomated production is the inevitable trend. The component automaticdetection has developed along with the unceasing development ofAutomation and the intellectualized technology in the modernizationindustrial production, there are great meanings that it is not only saving thetime and the work force, but also improving production efficiency and theprecision of the detection.The link-plate of chain is struck from the striking board, so the outline oflink-plate is possibly not entire;moreover, there are some differences in thecircumference and inner hole between the two surface of the link-plate, suchas the round angular zone, the luminous zone and the crack zone. Thisdifference owing to the processing means will affect the quality of theassembled chain very much. And in order to avoid this bad effect, andenhance the fatigue strength, the resistance to abrade and the appearancequality, each production enterprise will use the technical method generally,which is to distinguish the different faces of link-plate first, and then toassemble. This paper uses the computer image processing technology torealize the detection of the outline and automatic brand recognition oflink-plate. And we have done some research in the following aspects:Firstly we discuss the image preprocessing algorithms, mainly includingthe removing blur caused by uniform liner motion and the image smoothingprocessing. In latter, it compares and analyzes the complexity of algorithmsand the removing noise effect of both the neighborhood mean value filterand median filter in detail. And the experiment proves that the median filternot only could be able to remove the noise effectively, but also maintain theedge information of the link-plate image better, enhance the interestingcharacteristic in the picture. It was more advantageous to carry on theobservation to the goal object, but there are some shortage that it is notsuited to process the image including many thin line and the details.The edge in the image is one of the most basic and also the mostimportant characteristics of the image. The edge detection is classicalresearch topics in the computer vision and image processing domain all thetime, but people have not found a perfect solution to such question until now.The edge detection is one of basic steps of image processing, the imageanalysis, the image recognition, the computer vision as well as human vision.The result’s accuracy and reliability will affect the machine vision system tounderstand the objective world directly. First, it introduces the definition ofthe edge, the edge detection steps, requirements and the appraisal method.In the edge examination algorithm aspect, it mainly carries on thecomparison with all kinds of traditional algorithms (Roberts, Sobel, Prewitt,Laplacian, Krisch, canny and so on), elaborates the principle of eachalgorithm, and realizes them with the C + + Builder programming tool, andthen makes the analysis to the identical picture of static link-plate with thesemethods, compares the detection effect of each algorithm intuitionally.Regarding the fuzzy link-plate picture caused by the movement, anapproach to arc recognition based state machine was developed first time inthis paper. It has studied the arc recognition condition criterion, formulatedthe arc recognition state table and the next state operation table. Throughthe simulation with software programming, and the result shows that thismethod is practical. In this foundation, the arcs are grouped and matched tobe the shapes of link-plate, then we recognize the link-plate’s outlineaccording to the shape characteristic of link-plate, the mutual position ofcorresponding arc, the concave-convex relations. In order to distinguishbetween the obverse and the reverso of link-plate, we take the bland as thedetection object, through examining the edge quantity in the link joint waistregion whether bigger than the certain threshold value. If bigger, we thoughtthat we have found the bland in the above side, namely has found theobverse side that contains the bland.For the sake of fulfilling the real time request of image processing, thispaper discusses how to realize the link-plate detection and recognitionsystem with hardware, which is mainly composed of FPGA 1K50 and ARMLPC2106. It introduces the application and the hardware knowledge aboutFPGA 1K50 and the ARM LPC2106 simply, mainly discusses how to realizeand debug the median filter algorithm, the edge detection algorithm and thearc recognition based on state machine with FPGA. We defined thecommunication agreement between FPGA and ARM, and finishedcommunication procedure between FPGA and the ARM, the ARM’s startingcode and display procedure using serial interface.

Related Dissertations

  1. Research on Electromagnetic Acoustic Transducer Detection System Based on FPGA,TH878.2
  2. The Electric Power Steering System,U463.4
  3. Study on Drive and Control System of Five-Phase PMSM Based on FPGA,TM341
  4. Development of LXI Arbitrary Waveform Generator,TM935
  5. Design of Digital Predistorter for RF Power Amplifier Based on FPGA,TN722.75
  6. Synchronization and Channel Estimation Algorithm in Bursted OFDM System and Its Implementation on FPGA,TN919.3
  7. Performance Analysis of Anti-Multipath Capability in DSSS System and Study on the Compensation Methods,TN914.42
  8. Optimizing and Realising Research on Vedio Compression in TV Guidance System,TN919.81
  9. Research and Hardware Simulation of Multiuser Spread Spectrum Code Acquisition Based on FPGA,TN914.42
  10. Research on Basic Algorithms of Digital Image Processing and Implementation with FPGA,TP391.41
  11. Research of High Speed Image Pre-processing System Based on FPGA,TP391.41
  12. Research on Methods of Medical Ultrasound Image Denoising,TP391.41
  13. Development of Experimental Robot Control System Based on ARM,TP242.6
  14. Design of High-Speed Digital Image Acquisition and Interface Based on FPGA,TP274.2
  15. Development of Inductance Sensor Data Acquisition System Based on FPGA,TP274.2
  16. Data Collecting and Processing of Multi-Linear-CCD Visual Measuring System,TP274.2
  17. Research on Visual Servo System of Mechanical ARM,TP242.6
  18. Research on Software of TFT-LCD Testing Equipment Based on ARM,TN873.93
  19. Development of Serial Bus Analyzer Based on Nios,TP274
  20. Development of PMC High-speed Data Transmission Board Design Based on FPGA-RocketIO_X,TP274.2
  21. Development of Pxibus High Performance Digital I/O Module,TP274

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile