Marcelo-Chong's profile - activity

Dear ALl

Currently I am trying to detect a hand figure by using HAAR features

I train with the traincascade function provided by opencv

Now i would like to add SKIN color filter with the HAAR detection to be more accurate.

How can I combine this two method and comes out a detection ?

I am not sure I describe my question clearly, thanks.

I am using OpenCV 3 under linux ubunto 14

After i create the vec file, i use opencv_traincascade to train up my cascade.xml

but it always comes segmentation fault (core dump)

If I use LBP, it runs ok, but when i change to HAAR, it comes segmentation fault.

here's the result:

mac@ubuntu:~/Desktop$ opencv_traincascade -data data -vec cars.vec -bg negatives.txt -numPos 500 -numNeg 500 -w 20 -h 8 -numStage 5 -featureType HAAR -mode BASIC
PARAMETERS:
cascadeDirName: data
vecFileName: cars.vec
bgFileName: negatives.txt
numPos: 500
numNeg: 500
numStages: 20
precalcValBufSize[Mb] : 256
precalcIdxBufSize[Mb] : 256
stageType: BOOST
featureType: HAAR
sampleWidth: 20
sampleHeight: 8
boostType: GAB
minHitRate: 0.995
maxFalseAlarmRate: 0.5
weightTrimRate: 0.95
maxDepth: 1
maxWeakCount: 100
mode: BASIC

===== TRAINING 0-stage =====
<BEGIN
POS count : consumed   500 : 500
NEG count : acceptanceRatio    500 : 1
Precalculation time: 0
+----+---------+---------+
|  N |    HR   |    FA   |
+----+---------+---------+
|   1|        1|        1|
+----+---------+---------+
|   2|        1|        1|
+----+---------+---------+
|   3|        1|        1|
+----+---------+---------+
|   4|        1|        1|
+----+---------+---------+
|   5|        1|        1|
+----+---------+---------+
Segmentation fault (core dumped)

I try to set the pre cal memory to 1024, but it doesn't work too

can anyone help ? thx !!