Revision history [back]

Cascade Training Error OpenCV 3.0.0 - Train dataset for temp stage can not be filled. Branch training terminated. Cascade classifier can't be trained. Check the used training parameters.

"Train dataset for temp stage can not be filled. Branch training terminated. Cascade classifier can't be trained. Check the used training parameters."

I checked same problem under it.

・http://answers.opencv.org/question/91235/cascade-training-error-opencv-31-train-dataset-for-temp-stage-can-not-be-filled-branch-training-terminated-cascade-classifier-cant-be-trained-check/

However, error still occurred in this system. Also, I use OpenCV 3.0.0 with Python 2.7.12 on Ubuntu16.04.

#!/bin/sh

ls pos | xargs -I {} echo -e pos/{} > pslist.txt

paste pslist.txt plus.txt > pslist2.txt

mv pslist2.txt pslist.txt

ls neg | xargs -I {} echo -e neg/{} > nglist.txt

paste nglist.txt neg.txt > nglist2.txt

mv nglist2.txt nglist.txt

opencv_createsamples -info pslist.txt -num 99 -vec image.vec -w 24 -h 24

opencv_traincascade -data cascade -vec image.vec -bg nglist.txt -numPos 90 -numNeg 501 -numStages 20 -w 24 -h 24 -featureType LBP

pslist.txt

   pos/Flower100001.jpg 1 0 0 24 24
   pos/Flower100002.jpg 1 0 0 24 24
   pos/Flower100003.jpg 1 0 0 24 24
   pos/Flower100004.jpg 1 0 0 24 24
   pos/Flower100005.jpg 1 0 0 24 24
                ：

nglist.txt

   neg/Car100001.jpg    1 0 0 24 24
   neg/Car100002.jpg    1 0 0 24 24
   neg/Car100003.jpg    1 0 0 24 24
   neg/Car100004.jpg    1 0 0 24 24
   neg/Car100005.jpg    1 0 0 24 24
       　　　　　：

result

Info file name: pslist.txt
Img file name: (NULL)
Vec file name: image.vec
BG  file name: (NULL)
Num: 99
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...
Done. Created 99 samples
PARAMETERS:
cascadeDirName: cascade
vecFileName: image.vec
bgFileName: nglist.txt
numPos: 90
numNeg: 501
numStages: 20
precalcValBufSize[Mb] : 1024
precalcIdxBufSize[Mb] : 1024
acceptanceRatioBreakValue : -1
stageType: BOOST
featureType: LBP
sampleWidth: 24
sampleHeight: 24
boostType: GAB
minHitRate: 0.995
maxFalseAlarmRate: 0.5
weightTrimRate: 0.95
maxDepth: 1
maxWeakCount: 100

===== TRAINING 0-stage =====
<BEGIN
POS current samples: 1
POS current samples: 2
            ：
POS count : consumed   90 : 90
Train dataset for temp stage can not be filled. Branch training terminated.
Cascade classifier can't be trained. Check the used training parameters.

So, anyone teach me how to solve it?