A defect detection experimental system for microwave heated in-place recycling of asphalt pavements

In order to promote the intellectualization of the microwave hot recycling maintenance vehicle, an asphalt pavement defect detection system based on machine vision was applied to microwave heating recycling of asphalt pavements. A pavement defect image acquisition experimental system was built which including image acquisition, image preprocessing, defect objective extraction, distortion correction and image recognition. A weighted neighborhood average filtering algorithm was used to filter the image and the OTSU method combined with region growing method was employed to extract the defect object. The cubic polynomial method based on control object was used to correct barrel distortion and re-projection method was used to correct inclination distortion. In addition, the mapping relationship between defect-image-size with microwave-heater activated number was built to receive the controls parameters of maintenance vehicle. Experimental results demonstrate that the defect image detection system has better integration and reliability for developing efficient and flexible vehicle.