I'm trying to create custom haar classifier. I have opencv compiled on my machine.
I have 435 positive images and 4437 negatives (yeah, I know, probably waaaay to much), all of them grayscaled.
Now, my directory structure looks like this
opencv_createsamples
opencv_traincascade
positives.txt
negatives.txt
positives/
V-train001.ppm
...
negatives/
V-train001.ppm
...
negatives.txt looks like that:
./negatives/A-train0426.ppm
./negatives/Point-train1035.ppm
...
positives.txt:
positives/V-train362.ppm 1 1 1 66 76
positives/V-train227.ppm 1 1 1 66 76
positives/V-train154.ppm 1 1 1 65 76
Now, I'm creating samples with the following command:
novy@novy-K53SV:~/Downloads/opencv-3.1.0/build/bin$ ./opencv_createsamples -info positives.txt -vec sample.vec
Info file name: positives.txt
Img file name: (NULL)
Vec file name: sample.vec
BG file name: (NULL)
Num: 1000
BG color: 0
BG threshold: 80
Invert: FALSE
Max intensity deviation: 40
Max x angle: 1.1
Max y angle: 1.1
Max z angle: 0.5
Show samples: FALSE
Width: 24
Height: 24
Create training samples from images collection...
positives.txt(436) : parse errorDone. Created 435 samples
So it looks like samples have been created successfully.
However, when I'm trying to train my classifier I see the following:
novy@novy-K53SV:~/Downloads/opencv-3.1.0/build/bin$ ./opencv_traincascade -data haarcascade -vec sample.vec -bg negatives.txt -nstages 20 -nsplits 2 -minhitrate 0.999 -maxfalsealarm 0.5 -npos 200 -nneg 500
PARAMETERS:
cascadeDirName: haarcascade
vecFileName: sample.vec
bgFileName: negatives.txt
numPos: 2000
numNeg: 1000
numStages: 20
precalcValBufSize[Mb] : 1024
precalcIdxBufSize[Mb] : 1024
acceptanceRatioBreakValue : -1
stageType: BOOST
featureType: HAAR
sampleWidth: 24
sampleHeight: 24
boostType: GAB
minHitRate: 0.995
maxFalseAlarmRate: 0.5
weightTrimRate: 0.95
maxDepth: 1
maxWeakCount: 100
mode: BASIC
===== TRAINING 0-stage =====
↧
