A Novel Defect Inspection Approach Using Image Processing and Support Vector Machines in Bolts

To get rid of the weaknesses of the traditional defect inspection of the bolts, such as low efficiency, high cost, and poor-feasibility, a novel inspection approach for bolt looseness inspection based on the image processing technology using the charge coupled device (CCD) as the image sensor and the support vector machine (SVM) was proposed. The digital image of the tested bolt was collected by a DS-2AE7162-A CCD digital camera and input into computer by an image acquisition device. The original digital image with noise was processed by Gaussian filter and histogram equation to reduce its noise, and located by YCbCr color space. The samples feature descriptions of the bolts were extracted. A linear classifier model for defect inspection which based on the SVM was proposed. Experiments on several datasets demonstrate our approach is characterized by low cost, high efficiency, and easy realized.