Abstarct: One of important application of image processing is quality control in the production line of industries. In this thesis, for measuring the thickness of the cable insulation we employed image processing technique. For quality control of cables various parameters are considered which thickness of cable insulation is one of the most important parameters. There are international standards for criterions of the thickness of insulation for any type of cable. We studied measuring the thickness of three types cable: three-string, five-string and flat cables. According to standards, thickness of insulation in any cables should not be less than a level specified with precision one hundredth of millimeter. Various methods are currently used for these measurements that can be explained two sets of manual analysis of cable section to help projectors and using image processing techniques. In this thesis, techniques of image processing have been used to measure the thickness of the cable insulation. Imaging of cable insulation is done using a scanner with resolution of 1200 dpi. First images taken by image processing techniques are pre-processed. After acquiescing photograph, the first step is images binarization. Two methods have been used for binarizing of images. First we used Otsu to find a proper threshold to binarize the image, second we used images taken from the number of standard tools for determining threshold. To measure the parameters required in cable insulation that has been standardized, the first edges of the binarized image are extracted with using morphology operators and according to the cross-section topology of each cable, desired parameters are extracted as input to image processing and geometry of problem. Finally, the obtained parameters using the determined scale is converted to metric space. For evaluating proposed method, the results obtained from the proposed method for determining the threshold and Otsu thresholding algorithm and manual method have been compared together. The results show that the accuracy of the proposed algorithm is higher than Otsu thresholding algorithm. Furthermore, the proposed method in terms of time and labor make significant savings.