Bhavik's profile - activity

Whatever tutorial I an going through,the SVM classifier is used through .cpp code.However I am working on java, are there any tutorial through which I can implement SVM classifier through Java?

@thdrksdfthmn thanks for your advice,I will check and get back to you :)

Please tell me how to use this SVM classifier after eye detection??Sorry to ask such a stupid question,but I am a novice in this,so asking for your help.Or provide me with some links from where I can learn to to train a SVM classifier,and how to use it in code? Thanks in advance :)

Sorry to create a new thread , but I just need some suggestions.I need to write a java code with opencv to distinguish a person having closed eye fro a person having open eye.What is the best approach to be taken? 1)To train a haar classifier with closed eye and then implement it in the code? 2)use HoughCircles() method to detect the eye ball? I am asking this question because I have gone through some questions and answers in the forum and got to know that haar training is not giving a suitable output. If that is the answer what is to be done to implement the HoughCircle? Or are there any other solutions to this problem? Please Help Thanks in advance

yea it has been removed I guess

How to make the second classifier using HOG/SVM to cut out eye or eye+glass region and train the cascade.Although I have gone through http://answers.opencv.org/question/53... link and found some useful links about both SVM and HOG.But I couldn't figure out how to inplement it practically. It is far more comfusing as I couldnot run the hog.cpp in command prompt like the trainclassifier.exe.then how am I going to end up with the results Thanks in advance

How to make the second classifier using HOG/SVM to cut out eye or eye+glass region and train the cascade.Although I have gone through http://answers.opencv.org/question/53... link and found some useful links about both SVM and HOG.But I couldn't figure out how to inplement it practically. It is far more comfusing as I couldnot run the hog.cpp in command prompt like the trainclassifier.exe.then how am I going to end up with the results Thanks in advance

We are using 125 positive test images and 250 negative images. How many stages we should put to train the classifier? What if we give 1 stage only?

The trained cascade.xml is giving wrong result for the data which is being used for training the xml. why?

Are there any possible way to set those co-ordinates automatically.For training the classifier,we require huge number of images,and manually setting the co-ordinated of each image is time consuming process.Please provide a solution to ease the pain

ok thank you.

Yes I have seen,but couldn't find any proper answers for that questions.Thanks though.

ok,Thank u very much.I will try likewise.

I am having 25 positive images and 50 negative images,Running the traincascade.exe to generate the cascade classifier for 20 stages.But every time it is interrupting at 17th stage saying "Required leaf false alarm rate achieved. Branch training terminated."Even when I tried to run it for 10 stages it is getting interrupted at 5th step giving the same result.Why is it so???What is to be done to solve the problem???Please give solution. Although cascades.xml is generated by running the command opencv_traincascade -data cascade -vec text_classifier\positive.vec -bg negative_images\negative.dat -numPos 20 -numNeg 15 -numStag 1 -featureType HAAR -minHitRate 0.999 -maxFalseAlarmRate 0.5 -w 24 -h 30. Still many xmls are generated each for every successful stages,along with the param.xml and cascade.xml Yet I could not make which xml is to be used as an output ????

Suppose we want to train a classifier with human face wearing glasses.In this case while creating the .vec file do we need to specifically crop the eyes with glasses or the whole face along with the glasses?

Will be thankful to know the answer

How does the opencv_traincascade.exe identify the particular object within the image?Even if it can identify the approximate co-ordinates of the images by which we trained the classifiers,how is it going to identify the specified object.For eg:We want to train a classifier with eyes.Even though it can identify the coordinates where the eye resides within the image how is oit going too identify that it is nothing other than the eyes?Please explain Thanks in advance

We are creating .vec file of positive images through opencv_createsamples.exe where we are inputting the positive images within the -info command through .dat file, we are specifying co-ordinates for each of the images within the .dat file. Kindly elaborate on what those co-ordinates are and what is the purpose of putting them into .dat file?

According to my understanding the upper left corner is the 0,0 co-ordinate of the image. The data we put in the .dat file is like: image_path/image_name.jpg 1 250 180 350 200 here the 1 denotes the number of cropping we want to do from a single image 250 is the x co-ordinate value to the right side of 0,0 180 is the y co-ordinate value downwards to 0,0 350 denoted the width of the image 200 denotes the height of the image

Please let me know whether its correct or not and explain the same. Thanks in advance