| Sound transportation infrastructures are vital for economic development andsustainability. Pavement condition is a primary concern among agencies of thehighway infrastructures. Pavements are usually constructed with materials that exhibitdistresses due to various loading, environmental conditions, and aging. The earlydistresses are presented in the form of surface cracking mostly and threat the roadtraffic safety. Successful automation of surface distresses survey would reduce thecost of performing distresses survey and provide standardized and more objectiveresults for rehabilitation management.Nowadays, the image processing technology has been used in many disciplines.For example, the technology is being applied to automation, biological, medical andtransport engineering measurements. In transport engineering, this technology hasmainly used for accident detection, automatic recognition of license plates, automaticpavement crack detection, and pavement distress recognition.This research reviewed the advances in pavement distress detection equipmentsand pavement crack image processing technology. Based on linear CCD image sensor,this project set up pavement crack image collection device at the beginning, thedevice was able to grab pavement crack images in indoor virtual environmentsuccessfully. After that, the grayscale distribution characteristics of pavement crackimage subblocks and projection drawings on typical pavement cracks were analyzedfor detection and recognition. What’s more, the decision rules for pavement crackclassification was established to enable radon transform to distinguish four typicalpavement cracks. Last but not least, pavement crack image processing software wasdeveloped to realize the traditional and proposed algorithms.The verification results indicated that radon transform can effectively identify thetransverse crack, longitudinal crack, block crack and alligator crack, the precision was97.03%,98.85%,90.48%,73.33%respectively. |
PDF Full Text Request