For technical inspection of rubber products are essential methods of improving video scopes analyzing the degree of destruction and aging of rubber in an aggressive environment. The main factor determining the degree of destruction of the rubber product, the degree of coverage is cracked, which can be described as the amount of the total area, perimeter cracks, geometric shapes and other parameters. In the process of creating a methodology for assessing the degree of destruction of rubber products arises the problem of the development of machine vision algorithm for estimating the degree of coverage of the sample fractures and fracture characterization. For the development of image processing algorithm performed experimental studies on the artificial aging of several samples of products that are made from different rubbers. In the course of the experiments it was obtained several samples of shots vulcanizates in real time. To achieve the goals initially made light stabilization of array images using Gaussian filter. Thereafter, for each image binarization operation is applied. To highlight the contours of the surface damage of the sample is used Canny algorithm. The detected contours are converted into an array of pixels. However, a crack may be allocated to several contours. Therefore, an algorithm was developed by combining contours criterion of minimum distance between them. At the end of the calculation is made of the morphological features of each contour (area, perimeter, length, width, angle of inclination, the At the end of the calculation is made of the morphological features of each contour (area, perimeter, length, width, angle of inclination, the Minkowski dimension). Show schedule obtained by the method parameters destruction of samples of rubber products. The developed method allows you to automate assessment of the degree of aging of rubber products in telemetry systems, to study the dynamics of the aging process of polymers to obtain initial information for mathematical modeling of aging processes of polymers, to predict changes in the state of rubber products over time.

rubber products, machine vision algorithm, method, climate chamber

1. Li L., Sun L., Ning G., Tan S. Automatic pavement crack recognition based on BP neural networkю Promet – Traffic&Transportation, 2014, vol. 26, no. 1, pp. 11-22.

2. Broberg P., Runnemalm A. Detection of Surface Cracks in Welds using Active Thermography. 18th World Conference on Nondestructive Testing, Durban, South Africa. 2012, pp. 16-20.

3. 3Boonprakong P., Chamnongthai K. Bark Rubber Tree Crack Detection and Classification using Fractal Dimension. Proceedings of the 7th WSEAS International Conference on Robotics, Control & Manufacturing Technology, Hangzhou, China, April 15-17. 2007, pp. 229-232.

4. Gonsales R., Wuds R., Eddins S. Digital image processing using MATLAB. 1th ed. New York, Pearson Education, 2004. 620 p.

5. Bradsky G., Kaehler A. Learning OpenCV. O’Reilly, 2008. 580 p.

6. OpenCV 2.4.11.0 documentation. Available at: http://docs.opencv.org/.

7. Shuplntsov Yu.V., Ampilova N.B. Algorithm for computing the Minkowski dimension for halftone images. Izvestiya Rossiiskogo gosudarstvennogo pedagogicheskogo universiteta im. A.I. Gertsena [Proceedings of the Russian State Pedagogical University. A.I. Herzen], 2014, no. 165, pp. 99-106. (In Russ.)