opencv - Cascade training for road crack detection -


i've been working on cascade using lbp feature, me detect road crack on pavement pictures taken drone. did lot of test , different list of pictures, have lot of false positives.

im asking myself if method kind of detection , there better method use, other simple threshold?

training lbp or haar cascade classifiers detecting cracks on road in experience not best solution there no common denominator compared detecting faces etc. have proper set of data training? how many positives using? don't think there enough features selected , that's why getting many false positives, because cannot distinguish these during training.

start out different thresholds of road mentioned , see if can distinguish cracks. once visible enough in contrast road able go there.


Comments

Popular posts from this blog

Failed to execute goal org.apache.maven.plugins:maven-surefire-plugin:2.12:test (default-test) on project.Error occurred in starting fork -

windows - Debug iNetMgr.exe unhandle exception System.Management.Automation.CmdletInvocationException -

configurationsection - activeMq-5.13.3 setup configurations for wildfly 10.0.0 -