http://opencv.willowgarage.com/documentation/cpp/features2d_common_interfaces_of_descriptor_matchers.html#flannbasedmatcher
Please can somebody show me sample code or tell me how to use this class and methods.
I just want to match SURF's from a query image to those with an image set by applying Flann. I have seen many image match code in the samples but what still eludes me is a metric to quantify how similar an image is to other. Any help will be much appreciated.
Here's untested sample code
using namespace std;
using namespace cv;
Mat query; //the query image
vector<Mat> images; //set of images in your db
/* ... get the images from somewhere ... */
vector<vector<KeyPoint> > dbKeypoints;
vector<Mat> dbDescriptors;
vector<KeyPoint> queryKeypoints;
Mat queryDescriptors;
/* ... Extract the descriptors ... */
FlannBasedMatcher flannmatcher;
//train with descriptors from your db
flannmatcher.add(dbDescriptors);
flannmatcher.train();
vector<DMatch > matches;
flannmatcher.match(queryDescriptors, matches);
/* for kk=0 to matches.size()
the best match for queryKeypoints[matches[kk].queryIdx].pt
is dbKeypoints[matches[kk].imgIdx][matches[kk].trainIdx].pt
*/
Finding the most 'similar' image to the query image depends on your application. Perhaps the number of matched keypoints is adequate. Or you may need a more complex measure of similarity.
To reduce the number of false positives, you can compare the first most nearest neighbor to the second most nearest neighbor by taking the ratio of there distances.
distance(query,mostnearestneighbor)/distance(query,secondnearestneighbor) < T, the smaller the ratio is, the higher the distance of the second nearest neighbor to the query descriptor. This thus is a translation of high distinctiveness. Used in many computer vision papers that envision registration.
Related
I am not sure if this falls under the criteria of a proper question, but still, I would like to give it a shot.
I am looking for a library or function that takes two SIFT descriptors in a form of a file (or a matrix) of [number_of_keypoints][feature_0...feature_127] - meaning 128 features per file and allows comparison of images (I am using harris-affine alg. to extract them: http://www.robots.ox.ac.uk/~vgg/research/affine/det_eval_files/extract_features2.tar.gz ).
I am interested in a method that would allow me to find mutual nearest neighbours, that would accept number of keypoints in the neighbourhood and success ratio.
E.g.
Lets say I have two files with keypoints (described by SIFT descriptor) (image_1.sift, image_2.sift). I would like the method to accept: number of keypoints in the neighbourhood, match ratio, where match ratio means in pseudo code:
For each keypoint in image_1
Pick 50 nearest neighbours from image_1 -> List<KeyPoints> neighbours_1
For each keypoint in image_2
Pick 50 nearest neighbours from image_2 -> List<KeyPoints> neighbours_2
int numberOfMatches = 0;
foreach(neighbour in neighbours_1)
{
if(neighbour == neighbours_2.Find(neighbour))
numberOfMatches++;
}
The ratio is number of matches to number keypoints taken into consideration.
For example FindMutualKeypoints(image_1, image_2, 50, 0.7)
It can be c#, java, python or matlab implementation. I don't have much to do with image analysis on regular basis and before I start to write my own implementation, I assumed there probably is one out there already. I am having problem finding the correct terms in English from translation from my mother tongue (looks like the terms are quite different), which is probably the reason, why I could not find it yet.
I think openCV is the way to go.
Here is an example for it: link
It uses SURF descriptors, but you can also use SIFT.
You then call the FLANN matcher which also give you information about the quality of the matches.
I use SIFT to detect, describe feature points in two images as follows.
void FeaturePointMatching::SIFTFeatureMatchers(cv::Mat imgs[2], std::vector<cv::Point2f> fp[2])
{
cv::SiftFeatureDetector dec;
std::vector<cv::KeyPoint>kp1, kp2;
dec.detect(imgs[0], kp1);
dec.detect(imgs[1], kp2);
cv::SiftDescriptorExtractor ext;
cv::Mat desp1, desp2;
ext.compute(imgs[0], kp1, desp1);
ext.compute(imgs[1], kp2, desp2);
cv::BruteForceMatcher<cv::L2<float> > matcher;
std::vector<cv::DMatch> matches;
matcher.match(desp1, desp2, matches);
std::vector<cv::DMatch>::iterator iter;
fp[0].clear();
fp[1].clear();
for (iter = matches.begin(); iter != matches.end(); ++iter)
{
//if (iter->distance > 1000)
// continue;
fp[0].push_back(kp1.at(iter->queryIdx).pt);
fp[1].push_back(kp2.at(iter->trainIdx).pt);
}
// remove outliers
std::vector<uchar> mask;
cv::findFundamentalMat(fp[0], fp[1], cv::FM_RANSAC, 3, 1, mask);
std::vector<cv::Point2f> fp_refined[2];
for (size_t i = 0; i < mask.size(); ++i)
{
if (mask[i] != 0)
{
fp_refined[0].push_back(fp[0][i]);
fp_refined[1].push_back(fp[1][i]);
}
}
std::swap(fp_refined[0], fp[0]);
std::swap(fp_refined[1], fp[1]);
}
In the above code, I use findFundamentalMat() to remove outliers, but in the result img1 and img2 there are still some bad matches. In the images, each green line connects the matched feature point pair. And please ignore red marks. I can not find anything wrong, could anyone give me some hints? Thanks in advance.
RANSAC is just one of the robust estimators. In principle, one can use a variety of them but RANSAC has been shown to work quite well as long as your input data is not dominated by outliers. You can check other variants on RANSAC like MSAC, MLESAC, MAPSAC etc. which have some other interesting properties as well. You may find this CVPR presentation interesting (http://www.imgfsr.com/CVPR2011/Tutorial6/RANSAC_CVPR2011.pdf)
Depending on the quality of the input data, you can estimate the optimal number of RANSAC iterations as described here (https://en.wikipedia.org/wiki/RANSAC#Parameters)
Again, it is one of the robust estimator methods. You may take other statistical approaches like modelling your data with heavy tail distributions, trimmed least squares etc.
In your code you are missing the RANSAC step. RANSAC has basically 2 steps:
generate hypothesis (do a random selection of data points necessary to fit your mode: training data).
model evaluation (evaluate your model on the rest of the points: testing data)
iterate and choose the model that gives the lowest testing error.
RANSAC stands for RANdom SAmple Consensus, it does not remove outliers, it selects a group of points to calculate the fundamental matrix for that group of points. Then you need to do a re projection using the fundamental matrix just calculated with RANSAC to remove the outliers.
Like any algorithm, ransac is not perfect. You can try to run other disponible (in the opencv implementation) robust algorithms, like LMEDS. And you can reiterate, using the last points marked as inliers as input to a new estimation. And you can vary the threshold and confidence level. I suggest run ransac 1 ~ 3 times and then run LMEDS, that does not need a threshold, but only work well with at least +50% of inliers.
And, you can have geometrical order problems:
*If the baseline between the two stereo is too small the fundamental matrix estimation can be unreliable, and may be better to use findHomography() instead, for your purpose.
*if your images have some barrel/pincushin distortion, they are not in conformity with epipolar geometry and the fundamental matrix are not the correct mathematical model to link matches. In this case, you may have to calibrate your camera and then run undistort() and then process the output images.
i am using the following code in the descriptor_extractor_matcher.cpp sample to compute the descriptors of img1 (Mat descriptors01), write it to my disk and load it back (Mat descriptors1). (same steps for the keypoints, but code is rather much the same ...)
Ptr<DescriptorExtractor> descriptorExtractor = DescriptorExtractor::create( argv[2] );
...
Mat descriptors01;
descriptorExtractor->compute( img1, keypoints1, descriptors01 ); // compute descriptors
FileStorage storage("test.yml", FileStorage::WRITE); //save it to disc
storage << "blub" << descriptors01;
storage.release();
Mat descriptors1;
FileStorage storage1("test.yml", FileStorage::READ); // load it again
storage1["blub"] >> descriptors1;
storage1.release();
The keypoints & descriptors for image 2 are computed and used without saving and loading.
I am using only the loaded data (keypoints & descriptors) for image 1 for the matching, so for the descriptors: descriptors1.
Now here is the thing: if I compare the cases
A) Using the code above for computing, storing and loading;
B) Using only loaded data (without computing and store it again)
for the matching I get different results, as you can see in the pictures for keypoints aswell as for the matching descriptors. I would have expect no differences... What am I missing here? Must I compare 2 images, and cannot compare an image to a stored set of keypoints and it's descriptors ?
Of course I'm using the same values for [detectorType] [descriptorType] [matcherType] [matcherFilterType] [image1] [image2] [ransacReprojThreshold], by the way ;)
Thanks alot!
UPDATE:
It seems the issue is depending on the descriptor. Working with loaded descriptors works for SIFT and SURF, but not for ORB and other. Images: Results with different descriptors for case A and B:
Try repeating A or B individually and see if the results are coming out to be the same. I suspect they won't and I say that because, #1 Your object of interest has poor texture and that would result in poor descriptors. #2 The viewpoint change between the two images is huge and which leads to the problem of non-repeatability even for the best of the descriptors like SIFT.
Now, comes the part of how to solve this repeatability issue, #1 use some threshold on the norm of the descriptor so that only very strong features are used for matching. #2 use the epipolar constraint along with RANSAC to filter out wrong matches. I am attaching two images to show how the filter hugely affects the correspondences.
Using SURF to find correspondence between the two images (two images in red-cyan colormap)
After filtering the images using RANSAC using epipolar constraint.
Feel free to comment and discuss further over this issue. :-)
I have about 130,000 SIFT descriptors. I am building a hierarchical Kmeans-index using Opencv's flann module. After this I want to quantize these 130,000 descriptors (will quantize more later). I am using flann's knnsearch method for doing this. But the result of this method is something weird. For every descriptor the nearest index it is showing is the index of the descriptor itself. However, it should be displaying the cluster-ID of the nearest cluster which will be one of the leaves of the HIK-tree.
Should I try k=2
Here is a code snippet -
int k=1;
cv::flann::KMeansIndexParams indexParams(8,4,cvflann::FLANN_CENTERS_KMEANSPP) ;
cv::flann::Index hik_tree(cluster_data, indexParams);
Mat indices,dist;
hik_tree.knnSearch(cluster_data, indices, dist, k, cv::flann::SearchParams(64));
knnSearch is looking for the k-nearest neighbours in the index (it does not give the cluster-ID!). You build your index using cluster_data, and then you try to match cluster_data against itself. In this situation, it is not surprising that the closest neighbour to each descriptor is itself...
EDIT: If you want to get the centers, have a look at this (from the source of the FLANN library):
/**
* Chooses the initial centers using the algorithm proposed in the KMeans++ paper:
* Arthur, David; Vassilvitskii, Sergei - k-means++: The Advantages of Careful Seeding
*/
template <typename Distance>
class KMeansppCenterChooser : public CenterChooser<Distance>
{
...
k-NN is a supervised classification algorithm, that's why you are supposed to construct an Index object with your training samples, so use
cv::flann::Index hik_tree(samples, indexParams);
instead of
cv::flann::Index hik_tree(cluster_data, indexParams);
I need a simple and fast way to compare two images for similarity. I.e. I want to get a high value if they contain exactly the same thing but may have some slightly different background and may be moved / resized by a few pixel.
(More concrete, if that matters: The one picture is an icon and the other picture is a subarea of a screenshot and I want to know if that subarea is exactly the icon or not.)
I have OpenCV at hand but I am still not that used to it.
One possibility I thought about so far: Divide both pictures into 10x10 cells and for each of those 100 cells, compare the color histogram. Then I can set some made up threshold value and if the value I get is above that threshold, I assume that they are similar.
I haven't tried it yet how well that works but I guess it would be good enough. The images are already pretty much similar (in my use case), so I can use a pretty high threshold value.
I guess there are dozens of other possible solutions for this which would work more or less (as the task itself is quite simple as I only want to detect similarity if they are really very similar). What would you suggest?
There are a few very related / similar questions about obtaining a signature/fingerprint/hash from an image:
OpenCV / SURF How to generate a image hash / fingerprint / signature out of the descriptors?
Image fingerprint to compare similarity of many images
Near-Duplicate Image Detection
OpenCV: Fingerprint Image and Compare Against Database.
more, more, more, more, more, more, more
Also, I stumbled upon these implementations which have such functions to obtain a fingerprint:
pHash
imgSeek (GitHub repo) (GPL) based on the paper Fast Multiresolution Image Querying
image-match. Very similar to what I was searching for. Similar to pHash, based on An image signature for any kind of image, Goldberg et al. Uses Python and Elasticsearch.
iqdb
ImageHash. supports pHash.
Image Deduplicator (imagededup). Supports CNN, PHash, DHash, WHash, AHash.
Some discussions about perceptual image hashes: here
A bit offtopic: There exists many methods to create audio fingerprints. MusicBrainz, a web-service which provides fingerprint-based lookup for songs, has a good overview in their wiki. They are using AcoustID now. This is for finding exact (or mostly exact) matches. For finding similar matches (or if you only have some snippets or high noise), take a look at Echoprint. A related SO question is here. So it seems like this is solved for audio. All these solutions work quite good.
A somewhat more generic question about fuzzy search in general is here. E.g. there is locality-sensitive hashing and nearest neighbor search.
Can the screenshot or icon be transformed (scaled, rotated, skewed ...)? There are quite a few methods on top of my head that could possibly help you:
Simple euclidean distance as mentioned by #carlosdc (doesn't work with transformed images and you need a threshold).
(Normalized) Cross Correlation - a simple metrics which you can use for comparison of image areas. It's more robust than the simple euclidean distance but doesn't work on transformed images and you will again need a threshold.
Histogram comparison - if you use normalized histograms, this method works well and is not affected by affine transforms. The problem is determining the correct threshold. It is also very sensitive to color changes (brightness, contrast etc.). You can combine it with the previous two.
Detectors of salient points/areas - such as MSER (Maximally Stable Extremal Regions), SURF or SIFT. These are very robust algorithms and they might be too complicated for your simple task. Good thing is that you do not have to have an exact area with only one icon, these detectors are powerful enough to find the right match. A nice evaluation of these methods is in this paper: Local invariant feature detectors: a survey.
Most of these are already implemented in OpenCV - see for example the cvMatchTemplate method (uses histogram matching): http://dasl.mem.drexel.edu/~noahKuntz/openCVTut6.html. The salient point/area detectors are also available - see OpenCV Feature Detection.
I face the same issues recently, to solve this problem(simple and fast algorithm to compare two images) once and for all, I contribute an img_hash module to opencv_contrib, you can find the details from this link.
img_hash module provide six image hash algorithms, quite easy to use.
Codes example
origin lena
blur lena
resize lena
shift lena
#include <opencv2/core.hpp>
#include <opencv2/core/ocl.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/img_hash.hpp>
#include <opencv2/imgproc.hpp>
#include <iostream>
void compute(cv::Ptr<cv::img_hash::ImgHashBase> algo)
{
auto input = cv::imread("lena.png");
cv::Mat similar_img;
//detect similiar image after blur attack
cv::GaussianBlur(input, similar_img, {7,7}, 2, 2);
cv::imwrite("lena_blur.png", similar_img);
cv::Mat hash_input, hash_similar;
algo->compute(input, hash_input);
algo->compute(similar_img, hash_similar);
std::cout<<"gaussian blur attack : "<<
algo->compare(hash_input, hash_similar)<<std::endl;
//detect similar image after shift attack
similar_img.setTo(0);
input(cv::Rect(0,10, input.cols,input.rows-10)).
copyTo(similar_img(cv::Rect(0,0,input.cols,input.rows-10)));
cv::imwrite("lena_shift.png", similar_img);
algo->compute(similar_img, hash_similar);
std::cout<<"shift attack : "<<
algo->compare(hash_input, hash_similar)<<std::endl;
//detect similar image after resize
cv::resize(input, similar_img, {120, 40});
cv::imwrite("lena_resize.png", similar_img);
algo->compute(similar_img, hash_similar);
std::cout<<"resize attack : "<<
algo->compare(hash_input, hash_similar)<<std::endl;
}
int main()
{
using namespace cv::img_hash;
//disable opencl acceleration may(or may not) boost up speed of img_hash
cv::ocl::setUseOpenCL(false);
//if the value after compare <= 8, that means the images
//very similar to each other
compute(ColorMomentHash::create());
//there are other algorithms you can try out
//every algorithms have their pros and cons
compute(AverageHash::create());
compute(PHash::create());
compute(MarrHildrethHash::create());
compute(RadialVarianceHash::create());
//BlockMeanHash support mode 0 and mode 1, they associate to
//mode 1 and mode 2 of PHash library
compute(BlockMeanHash::create(0));
compute(BlockMeanHash::create(1));
}
In this case, ColorMomentHash give us best result
gaussian blur attack : 0.567521
shift attack : 0.229728
resize attack : 0.229358
Pros and cons of each algorithm
The performance of img_hash is good too
Speed comparison with PHash library(100 images from ukbench)
If you want to know the recommend thresholds for these algorithms, please check this post(http://qtandopencv.blogspot.my/2016/06/introduction-to-image-hash-module-of.html).
If you are interesting about how do I measure the performance of img_hash modules(include speed and different attacks), please check this link(http://qtandopencv.blogspot.my/2016/06/speed-up-image-hashing-of-opencvimghash.html).
Does the screenshot contain only the icon? If so, the L2 distance of the two images might suffice. If the L2 distance doesn't work, the next step is to try something simple and well established, like: Lucas-Kanade. Which I'm sure is available in OpenCV.
If you want to get an index about the similarity of the two pictures, I suggest you from the metrics the SSIM index. It is more consistent with the human eye. Here is an article about it: Structural Similarity Index
It is implemented in OpenCV too, and it can be accelerated with GPU: OpenCV SSIM with GPU
If you can be sure to have precise alignment of your template (the icon) to the testing region, then any old sum of pixel differences will work.
If the alignment is only going to be a tiny bit off, then you can low-pass both images with cv::GaussianBlur before finding the sum of pixel differences.
If the quality of the alignment is potentially poor then I would recommend either a Histogram of Oriented Gradients or one of OpenCV's convenient keypoint detection/descriptor algorithms (such as SIFT or SURF).
If for matching identical images - code for L2 distance
// Compare two images by getting the L2 error (square-root of sum of squared error).
double getSimilarity( const Mat A, const Mat B ) {
if ( A.rows > 0 && A.rows == B.rows && A.cols > 0 && A.cols == B.cols ) {
// Calculate the L2 relative error between images.
double errorL2 = norm( A, B, CV_L2 );
// Convert to a reasonable scale, since L2 error is summed across all pixels of the image.
double similarity = errorL2 / (double)( A.rows * A.cols );
return similarity;
}
else {
//Images have a different size
return 100000000.0; // Return a bad value
}
Fast. But not robust to changes in lighting/viewpoint etc.
Source
If you want to compare image for similarity,I suggest you to used OpenCV. In OpenCV, there are few feature matching and template matching. For feature matching, there are SURF, SIFT, FAST and so on detector. You can use this to detect, describe and then match the image. After that, you can use the specific index to find number of match between the two images.
Hu invariant moments is very powerful tool to compare two images
Hash functions are used in the undouble library to detect (near-)identical images (disclaimer: I am also the author). This is a simple and fast way to compare two or more images for similarity. It works using a multi-step process of pre-processing the images (grayscaling, normalizing, and scaling), computing the image hash, and the grouping of images based on a threshold value.