littletom's profile - activity

How can I load the existing svm in opencv3 by python?

yeah,I'm trying to extract the main information in order to reduce the noise and meaningless information in the pictures.As you told,if I can't do so or there are other ways to extract the main information using PCA?

I've trained an eigenvector matrix from 100 pictures through PCA algorithm.Then I tried to recover one of these 100 images using the matrix I gained,and succeed. It,however,was failure when I tried to recover those images other than in that 100 training samples.That is,information loss greatly.BTW,recovering in there means projecting a picture and then back project it.Please let me know if you are clear this time so I can edit the question again.

I've trained an eigenvector matrix from 100 pictures through PCA algorithm.Then I tried to recover one of these 100 images using the matrix I gained,and succeed. It,however,was failure when I tried to recover those images other than in that 100 training samples.That is,information loss greatly.BTW,recovering in there means projecting a picture and then back project it.

I'd like to visualize a image using different amount of eigenvectors,that is, I need to know what it looks like if the image have more or less features.

well,the output image doesn't change even if I've made an ajustment on the argument maxComponents. I'm sorry about that confusing anyone,and I am totally a greenhand in this filed.

import cv2 as cv
import numpy as np

img=cv.imread(r'T/s.jpg',cv.IMREAD_GRAYSCALE)
img=cv.resize(img,(512,512))
tdata=img.flatten()
tdata=tdata.reshape(512*512,1)
tdata=tdata-tdata.mean()
tdata=tdata/np.sqrt(tdata.var())
mean=np.array([])
mean,eigenvectors=cv.PCACompute(tdata,mean,cv.PCA_DATA_AS_COL,
                                maxComponents=9)
projectedVectors=cv.PCAProject(tdata,mean,eigenvectors)
result=cv.PCABackProject(tdata,mean,eigenvectors)
result=result.reshape(512,512)
cv.imshow('show',img)
cv.waitKey()
cv.imshow('show',result)
cv.waitKey()
cv.destroyAllWindows()

for example : PCACompute(data, mean[, eigenvectors[, maxComponents]]) in this documentation,what does these square brakets mean???

maybe you need a master,for example,the guy who'd already answered this question.

Although I've figured it out,thank you.

I need to save a lot of pictures in a "for" loop.I use python,imported opencv3.0.0 library already.

It works!You're amazing!

I mean reading many images at the same time so that I don't need to read them in one by one.Since I have to process these pictures using the same algorithm.

I don't know the way to load multiple images with python using OpenCV's API

look! guys,you are welcome!

as the title

as the title

well,it is not.

import cv2 as cv

imgpath='testImg.jpg'
img=cv.imread(imgpath,cv.IMREAD_GRAYSCALE)

orb=cv.ORB_create()
keypoints,descriptor=orb.detectAndCompute(img,None)

error: (-215) The data should normally be NULL! in function NumpyAllocator::allocate