The vision system is given a single training image (e.g. a piece of 2D artwork ) and it is asked whether the piece of artwork is present in the newly captured photos. The newly captured photos can contain a lot of any other object and when the artwork is presented, it must face up but may be occluded.
The pose space is x,y,rotation and scale. The artwork may be highly symmetric or not.
What is the latest state of the art handling this kind of problem?
I have tried/considered the following options but there are some problems in all of them. If my argument is invalid please correct me.
deep learning(rcnn/yolo): a lot of labeled data are needed means a lot of human labor is needed for each new pieces of artwork.
traditional machine learning(SVM,Random forest): same as above
sift/surf/orb + ransac or voting: when the artwork is symmetric, the features matched are mostly incorrect. A lot of time is needed in the ransac/voting stage.
generalized hough transform: the state space is too large for the voting table. Pyramid can be applied but it is difficult to choose some universal thresholds for different kinds of artwork to proceed down the pyramid.
chamfer matching: the state space is too large. Too much time is needed in searching across the state space.
Object detection requires a lot of labeled data of the same class to generalize well, and in your setting it would be impossible to train a network with only single instance.
I assume that in your case online object trackers can work, at least give it a try. There are some convolutional object trackers that work great like Siamese CNNs. The code is open source at github, and you can watch this video to see its performance.
Online object tracking: Given the initialized state (e.g., position
and size) of a target object in a frame of a video, the goal
of tracking is to estimate the states of the target in the subsequent
frames.-source-
You can try using traditional feature based image processing algorithm which might give true positive template matches up to a descent accuracy.
Given template image as in the question:
First dilate the image to join all very closely spaced connected
components.
Find the convex hull of the connected object obtained above,This will give you a polygon.
Use above polygon edge length information like (max-length/min-length) ratio as feature of the template.
Also find the pixel density in the polygon as second feature.
We have 2 features now.
Scene image feature vector:
Similarly Again in the scene image use dilation followed by connected components identification, define convex hull(polygon) around each connected objects and define feature vector for each object(edge info, pixel density).
Now as usual search for template feature vector in the scene image feature vectors data with minimum feature distance(also use certain upper level distance threshold value to avoid false positive matches).
This should give the true positive matches if available in the scene image.
Exception: This method would not work for occluded objects.
Related
I have an application where I have to detect the presence of some items in a scene. The items can be rotated and a little scaled (bigger or smaller). I've tried using keypoint detectors but they're not fast and accurate enough. So I've decided to first detect edges in the template and the search area, using Canny ( or a faster edge detection algo ), and then match the edges to find the position, orientation, and size of the match found.
All this needs to be done in less than a second.
I've tried using matchTemplate(), and matchShape() but the former is NOT scale and rotation invariant, and the latter doesn't work well with the actual images. Rotating the template image in order to match is also time consuming.
So far I have been able to detect the edges of the template but I don't know how to match them with the scene.
I've already gone through the following but wasn't able to get them to work (they're either using old version of OpenCV, or just not working with other images apart from those in the demo):
https://www.codeproject.com/Articles/99457/Edge-Based-Template-Matching
Angle and Scale Invariant template matching using OpenCV
https://answers.opencv.org/question/69738/object-detection-kinect-depth-images/
Can someone please suggest me an approach for this? Or a code snipped for the same if possible ?
This is my sample input image ( the parts to detect are marked in red )
These are some software that are doing this and also how I want it should be:
This topic is what I am actually dealing for a year on a project. So I will try to explain what my approach is and how I am doing that. I assume that you already did the preprocess steps(filters,brightness,exposure,calibration etc). And be sure you clean the noises on image.
Note: In my approach, I am collecting data from contours on a reference image which is my desired object. Then I am comparing these data with the other contours on the big image.
Use canny edge detection and find the contours on reference
image. You need to be sure here about that it shouldn't miss some parts of
contours. If it misses, probably preprocess part should have some
problems. The other important point is that you need to find an
appropriate mode of findContours because every modes have
different properties so you need to find an appropriate one for your
case. At the end you need to eliminate the contours which are okey
for you.
After getting contours from reference, you can find the length of
every contours using outputArray of findContours(). You can compare
these values on your big image and eliminate the contours which are
so different.
minAreaRect precisely draws a fitted, enclosing rectangle for
each contour. In my case, this function is very good to use. I am
getting 2 parameters using this function:
a) Calculate the short and long edge of fitted rectangle and compare the
values with the other contours on the big image.
b) Calculate the percentage of blackness or whiteness(if your image is
grayscale, get a percentage how many pixel close to white or black) and
compare at the end.
matchShape can be applied at the end to the rest of contours or you can also apply to all contours(I suggest first approach). Each contour is just an array so you can hold the reference contours in an array and compare them with the others at the end. After doing 3 steps and then applying matchShape is very good on my side.
I think matchTemplate is not good to use directly. I am drawing every contour to a different mat zero image(blank black surface) as a template image and then I compare with the others. Using a reference template image directly doesnt give good results.
OpenCV have some good algorithms about finding circles,convexity etc. If your situations are related with them, you can also use them as a step.
At the end, you just get the all data,values, and you can make a table in your mind. The rest is kind of statistical analysis.
Note: I think the most important part is preprocess part. So be sure about that you have a clean almost noiseless image and reference.
Note: Training can be a good solution for your case if you just want to know the objects exist or not. But if you are trying to do something for an industrial application, this is totally wrong way. I tried YOLO and haarcascade training algorithms several times and also trained some objects with them. The experiences which I get is that: they can find objects almost correctly but the center coordinates, rotation results etc. will not be totally correct even if your calibration is correct. On the other hand, training time and collecting data is painful.
You have rather bad image quality very bad light conditions, so you have only two ways:
1. To use filters -> binary threshold -> find_contours -> matchShape. But this very unstable algorithm for your object type and image quality. You will get a lot of wrong contours and its hard to filter them.
2. Haarcascades -> cut bounding box -> check the shape inside
All "special points/edge matching " algorithms will not work in such bad conditions.
I have created a point cloud of an irregular (non-planar) complex object using SfM. Each one of those 3D points was viewed in more than one image, so it has multiple (SIFT) features associated with it.
Now, I want to solve for the pose of this object in a new, different set of images using a PnP algorithm matching the features detected in the new images with the features associated with the 3D points in the point cloud.
So my question is: which descriptor do I associate with the 3D point to get the best results?
So far I've come up with a number of possible solutions...
Average all of the descriptors associated with the 3D point (taken from the SfM pipeline) and use that "mean descriptor" to do the matching in PnP. This approach seems a bit far-fetched to me - I don't know enough about feature descriptors (specifically SIFT) to comment on the merits and downfalls of this approach.
"Pin" all of the descriptors calculated during the SfM pipeline to their associated 3D point. During PnP, you would essentially have duplicate points to match with (one duplicate for each descriptor). This is obviously intensive.
Find the "central" viewpoint that the feature appears in (from the SfM pipeline) and use the descriptor from this view for PnP matching. So if the feature appears in images taken at -30, 10, and 40 degrees ( from surface normal), use the descriptor from the 10 degree image. This, to me, seems like the most promising solution.
Is there a standard way of doing this? I haven't been able to find any research or advice online regarding this question, so I'm really just curious if there is a best solution, or if it is dependent on the object/situation.
The descriptors that are used for matching in most SLAM or SFM systems are rotation and scale invariant (and to some extent, robust to intensity changes). That is why we are able to match them from different view points in the first place. So, in general it doesn't make much sense to try to use them all, average them, or use the ones from a particular image. If the matching in your SFM was done correctly, the descriptors of the reprojection of a 3d point from your point cloud in any of its observations should be very close, so you can use any of them 1.
Also, it seems to me that you are trying to directly match the 2d points to the 3d points. From a computational point of view, I think this is not a very good idea, because by matching 2d points with 3d ones, you lose the spatial information of the images and have to search for matches in a brute force manner. This in turn can introduce noise. But, if you do your matching from image to image and then propagate the results to the 3d points, you will be able to enforce priors (if you roughly know where you are, i.e. from an IMU, or if you know that your images are close), you can determine the neighborhood where you look for matches in your images, etc. Additionally, once you have computed your pose and refined it, you will need to add more points, no? How will you do it if you haven't done any 2d/2d matching, but just 2d/3d matching?
Now, the way to implement that usually depends on your application (how much covisibility or baseline you have between the poses from you SFM, etc). As an example, let's note your candidate image I_0, and let's note the images from your SFM I_1, ..., I_n. First, match between I_0 and I_1. Now, assume q_0 is a 2d point from I_0 that has successfully been matched to q_1 from I_1, which corresponds to some 3d point Q. Now, to ensure consistency, consider the reprojection of Q in I_2, and call it q_2. Match I_0 and I_2. Does the point to which q_0 is match in I_2 fall close to q_2? If yes, keep the 2d/3d match between q_0 and Q, and so on.
I don't have enough information about your data and your application, but I think that depending on your constraints (real-time or not, etc), you could come up with some variation of the above. The key idea anyway is, as I said previously, to try to match from frame to frame and then propagate to the 3d case.
Edit: Thank you for your clarifications in the comments. Here are a few thoughts (feel free to correct me):
Let's consider a SIFT descriptor s_0 from I_0, and let's note F(s_1,...,s_n) your aggregated descriptor (that can be an average or a concatenation of the SIFT descriptors s_i in their corresponding I_i, etc). Then when matching s_0 with F, you will only want to use a subset of the s_i that belong to images that have close viewpoints to I_0 (because of the 30deg problem that you mention, although I think it should be 50deg). That means that you have to attribute a weight to each s_i that depends on the pose of your query I_0. You obviously can't do that when constructing F, so you have to do it when matching. However, you don't have a strong prior on the pose (otherwise, I assume you wouldn't be needing PnP). As a result, you can't really determine this weight. Therefore I think there are two conclusions/options here:
SIFT descriptors are not adapted to the task. You can try coming up with a perspective-invariant descriptor. There is some literature on the subject.
Try to keep some visual information in the form of "Key-frames", as in many SLAM systems. It wouldn't make sense to keep all of your images anyway, just keep a few that are well distributed (pose-wise) in each area, and use those to propagate 2d matches to the 3d case.
If you only match between the 2d point of your query and 3d descriptors without any form of consistency check (as the one I proposed earlier), you will introduce a lot of noise...
tl;dr I would keep some images.
1 Since you say that you obtain your 3d reconstruction from an SFM pipline, some of them are probably considered inliers and some are outliers (indicated by a boolean flag). If they are outliers, just ignore them, if they are inliers, then they are the result of matching and triangulation, and their position has been refined multiple times, so you can trust any of their descriptors.
I'm using the ORB algorithm to detect and get the coordinates of the crossings of rope shown in the image, which is represented by the red dot. I want to detect the coordinates of the four points surrounding the crossing represented by the blue dots. All the four points have the same distance from the red spot.
Any idea how to get their coordinates by getting use of the red spot coordinate.
Thank you in Advance
Although you're using ORB, you're still going to need an algorithm to segment the rope from the background, or at least some technique to identify image chunks that belong to the rope and that are equidistant from the red dot. There are a number of options to explore.
It's important to consider your lighting & imaging as separate problems to be solved if this is meant to be a real-world application. This looks a bit like a problem for a class rather than for a application you'll sell and support, but you should still consider lighting:
Will your algorithm(s) still work when light level is reduced?
How will detection be affected by changes in camera pose relative to the surface where the rope will be located?
If you'll be detecting "black" rope, will the algorithm also be required to detect rope of different colors? dirty rope? rope on different backgrounds?
Since you're object of interest is rope, you have to consider a class of algorithms suitable for detection of non-rigid objects. Always consider the simplest solution first!
Connected Components
Connected components labeling is a traditional image processing algorithm and still suitable as the starting point for many applications. The last I knew, this was implemented in OpenCV as findContours(). This can also be called "blob finding" or some variant thereof.
https://en.wikipedia.org/wiki/Connected-component_labeling
https://docs.opencv.org/2.4/modules/imgproc/doc/structural_analysis_and_shape_descriptors.html?highlight=findcontours
Depending on lighting, you may have to take different steps to binarize the image before running connected components. As a start, convert the color image to grayscale, which will simplify the task significantly.
Try a manual threshold since you can quickly test a number of values to see the effect. Don't be too discouraged if the binarization isn't quite right--this can often be fixed with preprocessing.
If a range of manual thresholds works (e.g. 52 - 76 in an 8-bit grayscale range), then use an algorithm that will automatically calculate the threshold for you: Otsu, entropy-based methods, etc., will all offer comparable performance. Whichever technique works best, the code/algorithm can be tweaked further to optimize for your rope application.
If thresholding and binarization don't work--which for your rope application seems unlikely, at least how you've presented it--then switch to thinking in terms of gradient-based (edge-based, energy-based) techniques.
But assuming you can separate the rope from the background, you're still going to need a method to start at the red dot [within the rope] and move equal distances out to the blue points. More about that later after a discussion of other rope segmentation methods.
Note: connected components labeling can work in scenarios beyond just binarizing black & white images. If you can create a texture field or some other 2D representation of the image that makes it possible to distinguish the black rope from the relatively light background, you may be able to use a connected components algorithm. (Finding a "more complicated" or "more modern" algorithm isn't necessarily going to be the right approach.)
In a binarized image, blobs can be nested: on a white background you can have several black blobs, inside of one or more of which are white blobs, inside of which are black blobs, etc. An earlier version of OpenCV handled this reasonably well. (OpenCV is a nice starting point, and a touchpoint for many, but for a number of reasons it doesn't always compare favorably to other open source and commercial packages; popularity notwithstanding, OpenCV has some issues.)
Once you have a "blob" (a 4-connected region of pixels) in a 2D digital image, you can treat the blob as an object, at which point you have a number of options:
Edge tracing: trace around the inside and outside edges of the blob. From what I recall, OpenCV does (or at least should) have some relatively straightforward method to get the edges.
Split the blob into component blobs, each of which can be treated separately
Convert the blob to a polygon
...
A connected components algorithm should be high on the list of techniques to try if you have a non-rigid object.
Boolean Operations
Once you have the rope as a connected component (and possibly even without this), you can use boolean image operations to find the spots at the blue dots in your image:
Create a circular region in data, or even in the image
Find the intersection of the circle (an annulus) and the black region representing the rope. Using your original image, you should have four regions.
Find the center point of the intersection regions.
You could even try this without using connected components at all, but using connected components as part of the solution could make it more robust.
Polygon Simplification
If you have a blob, which in your application would be a connected set of black pixels representing the rope on the floor, then you can consider converting this blob to one or more polygons for further processing. There are advantages to working with polygons.
If you consider only the outside boundary of the rope, then you can see that the set of pixels defining the boundary represents a polygon. It's a polygon with a lot of points, and not a convex polygon, but a polygon nonetheless.
To simplify the polygon, you can use an algorithm such as Ramer-Douglas-Puecker:
https://en.wikipedia.org/wiki/Ramer%E2%80%93Douglas%E2%80%93Peucker_algorithm
Once you have a simplified polygon, you can try a few techniques to render useful data from the polygon
Angle Bisector Network
Triangulation (e.g. using ear clipping)
Triangulation is typically dependent on initial conditions, so the resulting triangulation for slighting different polygons (that is, rope -> blob -> polygon -> simplified polygon). So in your application it might be useful to triangulate the dark rope region, and then to connect the center of one triangle to the center of the next nearest triangle. You'll also have to deal with crossings, such as the rope overlap. Ultimately this can yield a "skeletonization" of the rope. Speaking of which...
Skeletonization
If the rope problem was posed to you as a class exercise, then it may have been a prompt to try skeletonization. You can read about it here:
https://en.wikipedia.org/wiki/Topological_skeleton
Skeletonization and thinning have their own problems to solve, but you should dig into them a bit and see those problems themselves.
The Medial Axis Transform (MAT) is a related concept. Long story there.
Edge-based techniques
There are a number of techniques to generate "edge images" based on edge strength, energy, entropy, etc. Making them robust takes a little effort. If you've had academic training in image processing you've likely heard of Harris, Sobel, Canny, and similar processing methods--none are magic bullets, but they're simple and dependable and will yield data you need.
An "edge image" consists of pixels representing the image gradient strength [and sometimes the gradient direction]. People may call this edge image something else, but it's the concept that matters.
What you then do with the edge data is another subject altogether. But one reason to think of edge images (or at least object borders) is that it reduces the amount of information your algorithm(s) will need to process.
Mean Shift (and related)
To get back to segmentation mentioned in the section on connected components, there are other methods for segmenting figures from a background: K-means, mean shift, and so on. You probably won't need any of those, but they're neat and worth studying.
Stroke Width Transform
This is an intriguing technique used to extract text from noisy backgrounds. Although it's intended for OCR, it could work for rope since the rope width is relatively constant, the rope shape varies, there are crossings, etc.
In short, and simplifying quite a bit, you can think of SWT as a means to find "strokes" (thick lines) by finding gradients antiparallel to each other. On either side of a stroke (or line), the edge gradient points normal to the object edge. The normal on one side of the stroke points opposite the direction of the normal on the other side of the stroke. By filtering for pixel-gradient pairs within a certain distance of each other, you can isolate certain strokes--even automatically. For your example the collection of points representing edge pairs for the rope would be much more common than other point pairs.
Non-Rigid Matching
There are techniques for matching non-rigid shapes, but they would not be worth exploring. If any of the techniques I mentioned above is unfamiliar to you, explore some of those first before you try any fancier algorithms.
CNNs, machine learning, etc.
Just don't even think of these methods as a starting point.
Other Considerations
If this were an application for industry, security, or whatnot, you'd have to determine how well your image processing worked under all environmental considerations. That's not an easy task, and can make all the difference between a setup that "works" in the lab and a setup that actually works in practice.
I hope that's of some help. Feel free to post a reply if I've confused more than helped, or if you want to explore some idea in more detail. Though I tried to touch on some common(ish) techniques, I didn't mention all the different ways of addressing this problem.
And briefly: once you have a skeleton, point network, or whatever representing a reduced data set for the rope and the red dot (the identified feature), a few techniques to find the items at the blue dots:
For a skeleton, trace along each "branch" of the rope outward from the know until the geodesic distance or straight-line 2D distance is the distance D that you want.
To use geometry, create a circle of width 1 - 2 pixels. Find the intersection of that circle and the rope. Find the center point of the intersections of circle and rope. (Also described above.)
Good luck!
I would like to know how I can use OpenCV to detect on my VideoCamera a Image. The Image can be one of 500 images.
What I'm doing at the moment:
- (void)viewDidLoad
{
[super viewDidLoad];
// Do any additional setup after loading the view.
self.videoCamera = [[CvVideoCamera alloc] initWithParentView:imageView];
self.videoCamera.delegate = self;
self.videoCamera.defaultAVCaptureDevicePosition = AVCaptureDevicePositionBack;
self.videoCamera.defaultAVCaptureSessionPreset = AVCaptureSessionPresetHigh;
self.videoCamera.defaultAVCaptureVideoOrientation = AVCaptureVideoOrientationPortrait;
self.videoCamera.defaultFPS = 30;
self.videoCamera.grayscaleMode = NO;
}
-(void)viewDidAppear:(BOOL)animated{
[super viewDidAppear:animated];
[self.videoCamera start];
}
#pragma mark - Protocol CvVideoCameraDelegate
#ifdef __cplusplus
- (void)processImage:(cv::Mat&)image;
{
// Do some OpenCV stuff with the image
cv::Mat image_copy;
cvtColor(image, image_copy, CV_BGRA2BGR);
// invert image
//bitwise_not(image_copy, image_copy);
//cvtColor(image_copy, image, CV_BGR2BGRA);
}
#endif
The images that I would like to detect are 2-5kb small. Few got text on them but others are just signs. Here a example:
Do you guys know how I can do that?
There are several things in here. I will break down your problem and point you towards some possible solutions.
Classification: Your main task consists on determining if a certain image belongs to a class. This problem by itself can be decomposed in several problems:
Feature Representation You need to decide how you are gonna model your feature, i.e. how are you going to represent each image in a feature space so you can train a classifier to separate those classes. The feature representation by itself is already a big design decision. One could (i) calculate the histogram of the images using n bins and train a classifier or (ii) you could choose a sequence of random patches comparison such as in a random forest. However, after the training, you need to evaluate the performance of your algorithm to see how good your decision was.
There is a known problem called overfitting, which is when you learn too well that you can not generalize your classifier. This can usually be avoided with cross-validation. If you are not familiar with the concept of false positive or false negative, take a look in this article.
Once you define your feature space, you need to choose an algorithm to train that data and this might be considered as your biggest decision. There are several algorithms coming out every day. To name a few of the classical ones: Naive Bayes, SVM, Random Forests, and more recently the community has obtained great results using Deep learning. Each one of those have their own specific usage (e.g. SVM ares great for binary classification) and you need to be familiar with the problem. You can start with simple assumptions such as independence between random variables and train a Naive Bayes classifier to try to separate your images.
Patches: Now you mentioned that you would like to recognize the images on your webcam. If you are going to print the images and display in a video, you need to handle several things. it is necessary to define patches on your big image (input from the webcam) in which you build a feature representation for each patch and classify in the same way you did in the previous step. For doing that, you could slide a window and classify all the patches to see if they belong to the negative class or to one of the positive ones. There are other alternatives.
Scale: Considering that you are able to detect the location of images in the big image and classify it, the next step is to relax the toy assumption of fixes scale. To handle a multiscale approach, you could image pyramid which pretty much allows you to perform the detection in multiresolution. Alternative approaches could consider keypoint detectors, such as SIFT and SURF. Inside SIFT, there is an image pyramid which allows the invariance.
Projection So far we assumed that you had images under orthographic projection, but most likely you will have slight perspective projections which will make the whole previous assumption fail. One naive solution for that would be for instance detect the corners of the white background of your image and rectify the image before building the feature vector for classification. If you used SIFT or SURF, you could design a way of avoiding explicitly handling that. Nevertheless, if your input is gonna be just squares patches, such as in ARToolkit, I would go for manual rectification.
I hope I might have given you a better picture of your problem.
I would recommend using SURF for that, because pictures can be on different distances form your camera, i.e changing the scale. I had one similar experiment and SURF worked just as expected. But SURF has very difficult adjustment (and expensive operations), you should try different setups before you get the needed results.
Here is a link: http://docs.opencv.org/modules/nonfree/doc/feature_detection.html
youtube video (in C#, but can give an idea): http://www.youtube.com/watch?v=zjxWpKCQqJc
I might not be qualified enough to answer this problem. Last time I seriously use OpenCV it was still 1.1. But just some thought on it, and hope it would help (currently I am interested in DIP and ML).
I think it will probably an easier task if you only need to classify an image, if the image is just one from (or very similar to) your 500 images. For this you could use SVM or some neural network (Felix already gave an excellent enumeration on that).
However, your problem seems to be that you need to first find this candidate image in your webcam, the location of which you have little clue beforehand. (let us know whether it is so. I think it is important.)
If so, the harder problem is the detection/localization of your candidate image.
I don't have a general solution for that. The first thing I would do is to see if there is some common feature in your 500 images (e.g., whether all of them enclosed by a red circle, or, half of them have circle and half of them have rectangle). If this can be done, the problem will be simpler (it would be similar to face detection problem, which have good solution).
In other words, this means that you first classify the 500 images to a few groups with common feature (by human), and detect the group first, then scale and use above mentioned technique to classify them into fine result. In this way, it will be more computationally acceptable than trying to detect 500 images one by one.
BTW, this ppt would help to give a visual clue of what is going on for feature extraction and image matching http://courses.cs.washington.edu/courses/cse455/09wi/Lects/lect6.pdf.
Detect vs recognize: detecting the image is just finding it on the background and from your comments I realized you may have your sings surrounded by the background. It might facilitate your algorithm if you can somehow crop your signs from the background (detect) before trying to recognize them. Recognizing is a next stage that presumes you can classify the cropped image correctly as the one seen before.
If you need real time speed and scale/rotation invariance neither SIFT no SURF will do this fast. Nowadays you can do much better if you shift the burden of image processing to a learning stage as was done by Lepitit. In short, he subjected each pattern to a bunch of affine transformations and trained a binary classification tree to recognize each point correctly by doing a lot of binary comparison tests. Trees are extremely fast and a way to go not to mention that most of the processing is done offline. This method is also more robust to off-plane rotations than SIFT or SURF. You will also learn about tree classification which may facilitate you last processing stage.
Finally a recognition stage is based not only on the number of matches but also on their geometric consistency. Since your signs look flat I suggest finding either affine or homography transformation that has most inliers when calculated between matched points.
Looking at your code though I realized that you may not follow any of these recommendations. It may be a good starting point for you to read about decision trees and then play with some sample code (see mushroom.cpp in the above mentioned link)
I am trying to do image detection in C++. I have two images:
Image Scene: 1024x786
Person: 36x49
And I need to identify this particular person from the scene. I've tried to use Correlation but the image is too noisy and therefore doesn't give correct/accurate results.
I've been thinking/researching methods that would best solve this task and these seem the most logical:
Gaussian filters
Convolution
FFT
Basically, I would like to move the noise around the images, so then I can use Correlation to find the person more effectively.
I understand that an FFT will be hard to implement and/or may be slow especially with the size of the image I'm using.
Could anyone offer any pointers to solving this? What would the best technique/algorithm be?
In Andrew Ng's Machine Learning class we did this exact problem using neural networks and a sliding window:
train a neural network to recognize the particular feature you're looking for using data with tags for what the images are, using a 36x49 window (or whatever other size you want).
for recognizing a new image, take the 36x49 rectangle and slide it across the image, testing at each location. When you move to a new location, move the window right by a certain number of pixels, call it the jump_size (say 5 pixels). When you reach the right-hand side of the image, go back to 0 and increment the y of your window by jump_size.
Neural networks are good for this because the noise isn't a huge issue: you don't need to remove it. It's also good because it can recognize images similar to ones it has seen before, but are slightly different (the face is at a different angle, the lighting is slightly different, etc.).
Of course, the downside is that you need the training data to do it. If you don't have a set of pre-tagged images then you might be out of luck - although if you have a Facebook account you can probably write a script to pull all of yours and your friends' tagged photos and use that.
A FFT does only make sense when you already have sort the image with kd-tree or a hierarchical tree. I would suggest to map the image 2d rgb values to a 1d curve and reducing some complexity before a frequency analysis.
I do not have an exact algorithm to propose because I have found that target detection method depend greatly on the specific situation. Instead, I have some tips and advices. Here is what I would suggest: find a specific characteristic of your target and design your code around it.
For example, if you have access to the color image, use the fact that Wally doesn't have much green and blue color. Subtract the average of blue and green from the red image, you'll have a much better starting point. (Apply the same operation on both the image and the target.) This will not work, though, if the noise is color-dependent (ie: is different on each color).
You could then use correlation on the transformed images with better result. The negative point of correlation is that it will work only with an exact cut-out of the first image... Not very useful if you need to find the target to help you find the target! Instead, I suppose that an averaged version of your target (a combination of many Wally pictures) would work up to some point.
My final advice: In my personal experience of working with noisy images, spectral analysis is usually a good thing because the noise tend to contaminate only one particular scale (which would hopefully be a different scale than Wally's!) In addition, correlation is mathematically equivalent to comparing the spectral characteristic of your image and the target.