Detecting Thread Features Based On Digital Image Processing Techniques

With the progress of computer technology, digital image processing technologygot great development, today digital image processing is one of the very populartechnology, it is widely used in medicine, astronomy, biology and national defencescience and other fields, there’s almost no nothing to do with the digital imageprocessing technology.Traditional thread detection method to be artificial, directly measuring accuracyhas closely relationship with operators, the entire measurement is accomplished by theoperating personnel, work efficiency is low, and many methods in the process ofdetection measuring instrument to direct contact with screw thread, the result is thatled to a decline in damage and accuracy of the thread. Caused by method of thelimitations of the traditional thread measuring precision is relatively low, the wholemeasuring process is time-consuming, low work efficiency. Therefore non-contactthread detection method based on image processing technology is a new kind of highprecision measurement, high efficiency and low cost.Through digital image processing refers to the computer for image denoising andenhancement, restoration, segmentation, feature extraction methods and techniques, etc.In simple terms, detection method based on digital image processing techniques towrite the test procedure of the thread parameters to the computer in advance, by thecamera thread images into the computer, through a series of image processing to detectthread parameters, to determine whether a thread is qualified and the whole testprocess without the participation of the completed by computer automatically to detectthread, its precision is higher. As we all know, the computer processing speed is veryfast, completed by computer for thread detection to a large extent improve theefficiency.The thesis introduces the important theory of image processing technology, andaccording to the characteristics of the thread contour algorithm design. First collection of original thread through my camera images, to get the thread of the original imagedigital preprocessing, including noise filtering, image enhancement, image thresholdtransform steps; Second local image image segmentation of interest, including edgedetection, contour extraction and refining process; Then in view of the thread placedthree direction, respectively in each case thread parameters of large diameter, paths,pitch and tooth type Angle algorithm.Finally, the output of thread parameters of large diameter, path, pitch and toothtype Angle calculation results, and measure the parameters of the screw thread formany times, to the error analysis of the results, analysis of the experimental systemerror and random error of the system.