Getting error while training my cascade classifier ?

3 次查看(过去 30 天)
I am trying to build my cascade classifier that can detect an area of image basically i want to get eyes from a face image. For this i am using image labeling. i have both positive and negative folder with more than 10 images Error: The total number of samples (both positive and negative together) must be greater than 10. You should add more positive samples, or increase 'NegativeSamplesFactor', or both.
Error in trainCascadeObjectDetector (line 173) featureParams, totalPosInstances] = populateTrainCascadeParams(parser);
Error in cascade (line 28) trainCascadeObjectDetector('eyes.xml',positiveInstances, ...

采纳的回答

Mabin Mariam Joseph
From the error snapshots attached, I see that the positives and negatives directories are not what is being specified by the attached code.
In the code it is : your_matlabroot\toolbox\vision\visiondata\positives and your_matlabroot\toolbox\vision\visiondata\negatives respectively which is different from the directories seen in the error snapshots.
Try the code after pointing to the correct directories.

更多回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Computer Vision Toolbox 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by