Font Size: a A A

Image Processing Of Asphalt Pavement Research Based On DSP

Posted on:2010-02-13Degree:MasterType:Thesis
Country:ChinaCandidate:S W WenFull Text:PDF
GTID:2178360278955717Subject:Traffic Information Engineering & Control
Abstract/Summary:PDF Full Text Request
The asphalt pavement surface crack is one of the main diseases of pavement surface in the highway, crack detection for the conservation and management of the road is very important information. For the detection of cracks information, the traditional methods based on PC by general-purpose CPU detection can not meet the requirements of more and more cracks detection in the road the development. Now Already production based on laser ranging and digital image processing multi-functional intelligent Pavement Surface Detection vehicles, because of the information contained in the asphalt pavement image complexity, image of the ever-increasing amount of data and other factors make detection results difficult to fully meet the actual demand. DSP (Digital Signal Processor) technology and digital image processing technology image processing system of asphalt pavement, its project objective is more rapid, more accurate and cost-effective detection.DSP technology of the system will be introduced to the asphalt pavement detection processed, using TI's TMS320C6416 as the imagery processing system's micro processing chip to the bituminous pavement crack image processing research. In the bituminous pavement image besides contains the road surface information which has us to need, has contained possibly also the image noise (for instance road surface oil mark, each sundry goods and so on), the breakage road surface (for instance crack, hollow and so on), in addition easily illumination, weather and other factor disturbance. This article overall evaluation bituminous pavement image and the DSP characteristic, realizes bituminous pavement image processing on DSP, and carries on the optimization to it.In DSP image data processing: First of all, the need for the road surface image using a special data structure stored images of gray-scale image data, the realization of the need to deal with image data compression, the DSP to reduce the amount of data to deal with the follow-up; Second, image enhanced processing, increase the use of two filtering binary segmentation image enhanced filtering, to achieve images of the cracks in asphalt pavement asphalt part of the enhanced information on the part of other non-interest in weakening the image has been very good to deal with; Finally, the use of edge detection and Enhanced Iterative threshold segmentation images. Asphalt pavement cracks after image pre-processing for surface cracks in the main parameter extraction, including cracks in the area, circumference, type, length, average width, the percentage of such cracks.Finally, the optimization process, the system software to optimize part of the call in addition to C6000 compiler optimization, software pipelining optimization properties of pure code-level optimization, but also the structure of the program a number of optimization, including scalability, clarity, Portable optimization, etc., which code for the development of follow-up maintenance work is very important, and very easy to implement, this system development method also can be easily transplanted to other image processing platform.The system can be applied to asphalt pavement cracks to automatically detect and be able to road management and maintenance of the road sector decision-making reference to scientific data detection.
Keywords/Search Tags:DSP, Digital Image Processing, Asphalt Pavement Surface, Crack, Pavement Surface Detection
PDF Full Text Request
Related items