Image Processing Algorithm Improvement for Coca-Cola Can Recognition

Question

One of the most interesting projects I ve worked on in the past couple of years was a project about image processing  The goal was to develop a system to be able to recognize Coca-Cola  cans   note that I m stressing the word  cans   you ll see why in a minute   You can see a sample below  with the can recognized in the green rectangle with scale and rotation     Some constraints on the project    The background could be very noisy  The can could have any scale or rotation or even orientation  within reasonable limits   The image could have some degree of fuzziness  contours might not be entirely straight   There could be Coca-Cola bottles in the image  and the algorithm should only detect the can  The brightness of the image could vary a lot  so you can t rely  too much  on color detection   The can could be partly hidden on the sides or the middle and possibly partly hidden behind a bottle  There could be no can at all in the image  in which case you had to find nothing and write a message saying so    So you could end up with tricky things like this  which in this case had my algorithm totally fail      I did this project a while ago  and had a lot of fun doing it  and I had a decent implementation  Here are some details about my implementation   Language  Done in C   using OpenCV library   Pre-processing  For the image pre-processing  i e  transforming the image into a more raw form to give to the algorithm  I used 2 methods    Changing color domain from RGB to HSV and filtering based on  red  hue  saturation above a certain threshold to avoid orange-like colors  and filtering of low value to avoid dark tones  The end result was a binary black and white image  where all white pixels would represent the pixels that match this threshold  Obviously there is still a lot of crap in the image  but this reduces the number of dimensions you have to work with     Noise filtering using median filtering  taking the median pixel value of all neighbors and replace the pixel by this value  to reduce noise  Using Canny Edge Detection Filter to get the contours of all items after 2 precedent steps     Algorithm  The algorithm itself I chose for this task was taken from this awesome book on feature extraction and called Generalized Hough Transform  pretty different from the regular Hough Transform   It basically says a few things    You can describe an object in space without knowing its analytical equation  which is the case here   It is resistant to image deformations such as scaling and rotation  as it will basically test your image for every combination of scale factor and rotation factor  It uses a base model  a template  that the algorithm will  learn   Each pixel remaining in the contour image will vote for another pixel which will supposedly be the center  in terms of gravity  of your object  based on what it learned from the model    In the end  you end up with a heat map of the votes  for example here all the pixels of the contour of the can will vote for its gravitational center  so you ll have a lot of votes in the same pixel corresponding to the center  and will see a peak in the heat map as below     Once you have that  a simple threshold-based heuristic can give you the location of the center pixel  from which you can derive the scale and rotation and then plot your little rectangle around it  final scale and rotation factor will obviously be relative to your original template   In theory at least     Results  Now  while this approach worked in the basic cases  it was severely lacking in some areas    It is extremely slow  I m not stressing this enough  Almost a full day was needed to process the 30 test images  obviously because I had a very high scaling factor for rotation and translation  since some of the cans were very small  It was completely lost when bottles were in the image  and for some reason almost always found the bottle instead of the can  perhaps because bottles were bigger  thus had more pixels  thus more votes  Fuzzy images were also no good  since the votes ended up in pixel at random locations around the center  thus ending with a very noisy heat map  In-variance in translation and rotation was achieved  but not in orientation  meaning that a can that was not directly facing the camera objective wasn t recognized    Can you help me improve my specific algorithm  using exclusively OpenCV features  to resolve the four specific issues mentioned   I hope some people will also learn something out of it as well  after all I think not only people who ask questions should learn

User · Answer

Fun problem  when I glanced at your bottle image I thought it was a can too  But  as a human  what I did to tell the difference is that I then noticed it was also a bottle     So  to tell cans and bottles apart  how about simply scanning for bottles first  If you find one  mask out the label before looking for cans   Not too hard to implement if you re already doing cans  The real downside is it doubles your processing time   But thinking ahead to real-world applications  you re going to end up wanting to do bottles anyway  -

User · Answer

I like the challenge and wanted to give an answer  which solves the issue  I think     Extract features  keypoints  descriptors such as SIFT  SURF  of the logo  Match the points with a model image of the logo  using Matcher such as Brute Force   Estimate the coordinates of the rigid body  PnP problem - SolvePnP  Estimate the cap position according to the rigid body  Do back-projection and calculate the image pixel position  ROI  of the cap of the bottle  I assume you have the intrinsic parameters of the camera  Check with a method whether the cap is there or not  If there  then this is the bottle    Detection of the cap is another issue  It can be either complicated or simple  If I were you  I would simply check the color histogram in the ROI for a simple decision     Please  give the feedback if I am wrong  Thanks

User · Answer

If you are interested in it being realtime  then what you need is to add in a pre-processing filter to determine what gets scanned with the heavy-duty stuff  A good fast  very real time  pre-processing filter that will allow you to scan things that are more likely to be a coca-cola can than not before moving onto more iffy things is something like this  search the image for the biggest patches of color that are a certain tolerance away from the sqrt pow red 2    pow blue 2    pow green 2   of your coca-cola can  Start with a very strict color tolerance  and work your way down to more lenient color tolerances  Then  when your robot runs out of an allotted time to process the current frame  it uses the currently found bottles for your purposes  Please note that you will have to tweak the RGB colors in the sqrt pow red 2    pow blue 2    pow green 2   to get them just right   Also  this is gona seem really dumb  but did you make sure to turn on -oFast compiler optimizations when you compiled your C code

User · Answer

Isn t it difficult even for humans to distinguish between a bottle and a can in the second image  provided the transparent region of the bottle is hidden    They are almost the same except for a very small region  that is  width at the top of the can is a little small while the wrapper of the bottle is the same width throughout  but a minor change right    The first thing that came to my mind was to check for the red top of bottle  But it is still a problem  if there is no top for the bottle  or if it is partially hidden  as mentioned above    The second thing I thought was about the transparency of bottle  OpenCV has some works on finding transparent objects in an image  Check the below links    OpenCV Meeting Notes Minutes 2012-03-19 OpenCV Meeting Notes Minutes 2012-02-28   Particularly look at this to see how accurately they detect glass    OpenCV Meeting Notes Minutes 2012-04-24   See their implementation result     They say it is the implementation of the paper  A Geodesic Active Contour Framework for Finding Glass  by K  McHenry and J  Ponce  CVPR 2006   It might be helpful in your case a little bit  but problem arises again if the bottle is filled   So I think here  you can search for the transparent body of the bottles first or for a red region connected to two transparent objects laterally which is obviously the bottle   When working ideally  an image as follows      Now you can remove the yellow region  that is  the label of the bottle and run your algorithm to find the can   Anyway  this solution also has different problems like in the other solutions    It works only if your bottle is empty  In that case  you will have to search for the red region between the two black colors  if the Coca Cola liquid is black   Another problem if transparent part is covered    But anyway  if there are none of the above problems in the pictures  this seems be to a better way

User · Answer

If you are not limited to just a camera which wasn t in one of your constraints perhaps you can move to using a range sensor like the Xbox Kinect  With this you can perform depth and colour based matched segmentation of the image  This allows for faster separation of objects in the image  You can then use ICP matching or similar techniques to even match the shape of the can rather then just its outline or colour and given that it is cylindrical this may be a valid option for any orientation if you have a previous 3D scan of the target  These techniques are often quite quick especially when used for such a specific purpose which should solve your speed problem    Also I could suggest  not necessarily for accuracy or speed but for fun you could use a trained neural network on your hue segmented image to identify the shape of the can  These are very fast and can often be up to 80 90  accurate  Training would be a little bit of a long process though as you would have to manually identify the can in each image

User · Answer

I really like Darren Cook s and stacker s answers to this problem  I was in the midst of throwing my thoughts into a comment on those  but I believe my approach is too answer-shaped to not leave here   In short summary  you ve identified an algorithm to determine that a Coca-Cola logo is present at a particular location in space  You re now trying to determine  for arbitrary orientations and arbitrary scaling factors  a heuristic suitable for distinguishing Coca-Cola cans from other objects  inclusive of  bottles  billboards  advertisements  and Coca-Cola paraphernalia all associated with this iconic logo  You didn t call out many of these additional cases in your problem statement  but I feel they re vital to the success of your algorithm   The secret here is determining what visual features a can contains or  through the negative space  what features are present for other Coke products that are not present for cans  To that end  the current top answer sketches out a basic approach for selecting  can  if and only if  bottle  is not identified  either by the presence of a bottle cap  liquid  or other similar visual heuristics   The problem is this breaks down  A bottle could  for example  be empty and lack the presence of a cap  leading to a false positive  Or  it could be a partial bottle with additional features mangled  leading again to false detection  Needless to say  this isn t elegant  nor is it effective for our purposes   To this end  the most correct selection criteria for cans appear to be the following    Is the shape of the object silhouette  as you sketched out in your question  correct  If so   1  If we assume the presence of natural or artificial light  do we detect a chrome outline to the bottle that signifies whether this is made of aluminum  If so   1  Do we determine that the specular properties of the object are correct  relative to our light sources  illustrative video link on light source detection   If so   1  Can we determine any other properties about the object that identify it as a can  including  but not limited to  the topological image skew of the logo  the orientation of the object  the juxtaposition of the object  for example  on a planar surface like a table or in the context of other cans   and the presence of a pull tab  If so  for each   1    Your classification might then look like the following    For each candidate match  if the presence of a Coca Cola logo was detected  draw a gray border  For each match over  2  draw a red border    This visually highlights to the user what was detected  emphasizing weak positives that may  correctly  be detected as mangled cans   The detection of each property carries a very different time and space complexity  and for each approach  a quick pass through http   dsp stackexchange com is more than reasonable for determining the most correct and most efficient algorithm for your purposes  My intent here is  purely and simply  to emphasize that detecting if something is a can by invalidating a small portion of the candidate detection space isn t the most robust or effective solution to this problem  and ideally  you should take the appropriate actions accordingly   And hey  congrats on the Hacker News posting  On the whole  this is a pretty terrific question worthy of the publicity it received

User · Answer

There is a computer vision package called HALCON from MVTec whose demos could give you good algorithm ideas  There is plenty of examples similar to your problem that you could run in demo mode and then look at the operators in the code and see how to implement them from existing OpenCV operators    I have used this package to quickly prototype complex algorithms for problems like this and then find how to implement them using existing OpenCV features  In particular for your case you could try to implement in OpenCV the functionality embedded in the operator find scaled shape model  Some operators point to the scientific paper regarding algorithm implementation which can help to find out how to do something similar in OpenCV  Hope this helps

User · Answer

Maybe too many years late  but nevertheless a theory to try   The ratio of bounding rectangle of red logo region to the overall dimension of the bottle can is different  In the case of Can  should be 1 1  whereas will be different in that of bottle  with or without cap    This should make it easy to distinguish between the two   Update  The horizontal curvature of the logo region will be different between the Can and Bottle due their respective size difference  This could be specifically useful if your robot needs to pick up can bottle  and you decide the grip accordingly

User · Answer

Please take a look at Zdenek Kalal s Predator tracker  It requires some training  but it can actively learn how the tracked object looks at different orientations and scales and does it in realtime   The source code is available on his site  It s in MATLAB  but perhaps there is a Java implementation already done by a community member  I have succesfully re-implemented the tracker part of TLD in C   If I remember correctly  TLD is using Ferns as the keypoint detector  I use either SURF or SIFT instead  already suggested by  stacker  to reacquire the object if it was lost by the tracker  The tracker s feedback makes it easy to build with time a dynamic list of sift surf templates that with time enable reacquiring the object with very high precision   If you re interested in my C  implementation of the tracker  feel free to ask

User · Answer

To speed things up  I would take advantage of the fact that you are not asked to find an arbitrary image object  but specifically one with the Coca-Cola logo  This is significant because this logo is very distinctive  and it should have a characteristic  scale-invariant signature in the frequency domain  particularly in the red channel of RGB  That is to say  the alternating pattern of red-to-white-to-red encountered by a horizontal scan line  trained on a horizontally aligned logo  will have a distinctive  rhythm  as it passes through the central axis of the logo  That rhythm will  speed up  or  slow down  at different scales and orientations  but will remain proportionally equivalent  You could identify define a few dozen such scanlines  both horizontally and vertically through the logo and several more diagonally  in a starburst pattern  Call these the  signature scan lines      Searching for this signature in the target image is a simple matter of scanning the image in horizontal strips  Look for a high-frequency in the red-channel  indicating moving from a red region to a white one   and once found  see if it is followed by one of the frequency rhythms identified in the training session  Once a match is found  you will instantly know the scan-line s orientation and location in the logo  if you keep track of those things during training   so identifying the boundaries of the logo from there is trivial    I would be surprised if this weren t a linearly-efficient algorithm  or nearly so  It obviously doesn t address your can-bottle discrimination  but at least you ll have your logos     Update  for bottle recognition I would look for coke  the brown liquid  adjacent to the logo --  that is  inside the bottle  Or  in the case of an empty bottle  I would look for a cap which will always have the same basic shape  size  and distance from the logo and will typically be all white or red  Search for a solid color eliptical shape where a cap should be  relative to the logo  Not foolproof of course  but your goal here should be to find the easy ones fast     It s been a few years since my image processing days  so I kept this suggestion high-level and conceptual  I think it might slightly approximate how a human eye might operate -- or at least how my brain does

User · Answer

I m not aware of OpenCV but looking at the problem logically I think you could differentiate between bottle and can by changing the image which you are looking for i e  Coca Cola  You should incorporate till top portion of can as in case of can there is silver lining at top of coca cola and in case of bottle there will be no such silver lining   But obviously this algorithm will fail in cases where top of can is hidden  but in such case even human will not be able to differentiate between the two  if only coca cola portion of bottle can is visible

User · Answer

I would detect red rectangles  RGB -  HSV  filter red -  binary image  close  dilate then erode  known as imclose in matlab   Then look through rectangles from largest to smallest  Rectangles that have smaller rectangles in a known position scale can both be removed  assuming bottle proportions are constant  the smaller rectangle would be a bottle cap     This would leave you with red rectangles  then you ll need to somehow detect the logos to tell if they re a red rectangle or a coke can  Like OCR  but with a known logo

User · Answer

Looking at shape  Take a gander at the shape of the red portion of the can bottle  Notice how the can tapers off slightly at the very top whereas the bottle label is straight  You can distinguish between these two by comparing the width of the red portion across the length of it   Looking at highlights  One way to distinguish between bottles and cans is the material  A bottle is made of plastic whereas a can is made of aluminum metal  In sufficiently well-lit situations  looking at the specularity would be one way of telling a bottle label from a can label   As far as I can tell  that is how a human would tell the difference between the two types of labels  If the lighting conditions are poor  there is bound to be some uncertainty in distinguishing the two anyways  In that case  you would have to be able to detect the presence of the transparent translucent bottle itself

User · Answer

There are a bunch of color descriptors used to recognise objects  the paper below compares a lot of them  They are specially powerful when combined with SIFT or SURF  SURF or SIFT alone are not very useful in a coca cola can image because they don t recognise a lot of interest points  you need the color information to help  I use BIC  Border Interior Pixel Classi cation  with SURF in a project and it worked great to recognise objects   Color descriptors for Web image retrieval  a comparative study

User · Answer

Am a few years late in answering this question  With the state of the art pushed to its limits by CNNs in the last 5 years I wouldn t use OpenCV to do this task now   I know you specifically wanted OpenCv features in the question  I feel object detection algorithms such as Faster-RCNNs  YOLO  SSD etc would ace this problem with a significant margin compared to OpenCV features  If I were to tackle this problem now  after 6 years     I would definitely use Faster-RCNN

User · Answer

The first things I would look for are color - like RED   when doing Red eye detection in an image - there is a certain color range to detect   some characteristics about it considering the surrounding area and such as distance apart from the other eye if it is indeed visible in the image   1  First characteristic is color and Red is very dominant  After detecting the Coca Cola Red there are several items of interest  1A  How big is this red area  is it of sufficient quantity to make a determination of a true can or not - 10 pixels is probably not enough   1B  Does it contain the color of the Label -  Coca-Cola   or wave   1B1  Is there enough to consider a high probability that it is a label   Item 1 is kind of a short cut - pre-process if that doe snot exist in the image - move on   So if that is the case I can then utilize that segment of my image and start looking more zoom out of the area in question a little bit - basically look at the surrounding region   edges     2  Given the above image area ID d in 1 - verify the surrounding points  edges  of the item in question  A  Is there what appears to be a can top or bottom - silver  B  A bottle might appear transparent   but so might a glass table - so is there a glass table shelf or a transparent area - if so there are multiple possible out comes  A Bottle MIGHT have a red cap  it might not  but it should have either the shape of the bottle top   thread screws  or a cap   C  Even if this fails A and B it still can be a can - partial   This is more complex when it is partial because a partial bottle   partial can might look the same   so some more processing of measurement of the Red region edge to edge   small bottle might be similar in size     3  After the above analysis that is when I would look at the lettering and the wave logo - because I can orient my search for some of the letters in the words As you might not have all of the text due to not having all of the can  the wave would align at certain points to the text  distance wise  so I could search for that probability and know which letters should exist at that point of the wave at distance x

User · Answer

You need a program that learns and improves classification accuracy organically from experience    I ll suggest deep learning  with deep learning this becomes a trivial problem    You can retrain the inception v3 model on Tensorflow   How to Retrain Inception s Final Layer for New Categories   In this case  you will be training a convolutional neural network to classify an object as either a coca-cola can or not

User · Answer

This may be a very naive idea  or may not work at all   but the dimensions of all the coke cans are fixed  So may be if the same image contains both a can and a bottle then you can tell them apart by size considerations  bottles are going to be larger   Now because of missing depth  i e  3D mapping to 2D mapping  its possible that a bottle may appear shrunk and there isn t a size difference  You may recover some depth information using stereo-imaging and then recover the original size

User · Answer

As alternative to all these nice solutions  you can train your own classifier and make your application robust to errors  As example  you can use Haar Training  providing a good number of positive and negative images of your target   It can be useful to extract only cans and can be combined with the detection of transparent objects

User · Answer

Deep Learning Gather at least a few hundred images containing cola cans  annotate the bounding box around them as positive classes  include cola bottles and other cola products label them negative classes as well as random objects  Unless you collect a very large dataset  perform the trick of using deep learning features for small dataset  Ideally using a combination of Support Vector Machines SVM  with deep neural nets  Once you feed the images to a previously trained deep learning model e g  GoogleNet   instead of using neural network s decision  final  layer to do classifications  use previous layer s   data as features to train your classifier  OpenCV and Google Net  http   docs opencv org trunk d5 de7 tutorial dnn googlenet html OpenCV and SVM  http   docs opencv org 2 4 doc tutorials ml introduction to svm introduction to svm html

User · Answer

An alternative approach would be to extract features  keypoints  using the scale-invariant feature transform  SIFT  or Speeded Up Robust Features  SURF   You can find a nice OpenCV code example in Java  C    and Python on this page  Features2D   Homography to find a known object Both algorithms are invariant to scaling and rotation  Since they work with features  you can also handle occlusion  as long as enough keypoints are visible    Image source  tutorial example The processing takes a few hundred ms for SIFT  SURF is bit faster  but it not suitable for real-time applications  ORB uses FAST which is weaker regarding rotation invariance  The original papers  SURF  Speeded Up Robust Features Distinctive Image Features from Scale-Invariant Keypoints ORB  an efficient alternative to SIFT or SURF

User · Answer

I like your question  regardless of whether it s off topic or not  P  An interesting aside  I ve just completed a subject in my degree where we covered robotics and computer vision  Our project for the semester was incredibly similar to the one you describe   We had to develop a robot that used an Xbox Kinect to detect coke bottles and cans on any orientation in a variety of lighting and environmental conditions  Our solution involved using a band pass filter on the Hue channel in combination with the hough circle transform  We were able to constrain the environment a bit  we could chose where and how to position the robot and Kinect sensor   otherwise we were going to use the SIFT or SURF transforms   You can read about our approach on my blog post on the topic

User · Answer

Hmm  I actually think I m onto something  this is like the most interesting question ever - so it d be a shame not to continue trying to find the  perfect  answer  even though an acceptable one has been found      Once you find the logo  your troubles are half done  Then you only have to figure out the differences between what s around the logo  Additionally  we want to do as little extra as possible  I think this is actually this easy part     What is around the logo  For a can  we can see metal  which despite the effects of lighting  does not change whatsoever in its basic colour  As long as we know the angle of the label  we can tell what s directly above it  so we re looking at the difference between these     Here  what s above and below the logo is completely dark  consistent in colour  Relatively easy in that respect     Here  what s above and below is light  but still consistent in colour  It s all-silver  and all-silver metal actually seems pretty rare  as well as silver colours in general  Additionally  it s in a thin slither and close enough to the red that has already been identified so you could trace its shape for its entire length to calculate a percentage of what can be considered the metal ring of the can  Really  you only need a small fraction of that anywhere along the can to tell it is part of it  but you still need to find a balance that ensures it s not just an empty bottle with something metal behind it     And finally  the tricky one  But not so tricky  once we re only going by what we can see directly above  and below  the red wrapper  Its transparent  which means it will show whatever is behind it  That s good  because things that are behind it aren t likely to be as consistent in colour as the silver circular metal of the can  There could be many different things behind it  which would tell us that it s an empty  or filled with clear liquid  bottle  or a consistent colour  which could either mean that it s filled with liquid or that the bottle is simply in front of a solid colour  We re working with what s closest to the top and bottom  and the chances of the right colours being in the right place are relatively slim  We know it s a bottle  because it hasn t got that key visual element of the can  which is relatively simplistic compared to what could be behind a bottle    that last one was the best I could find of an empty large coca cola bottle - interestingly the cap AND ring are yellow  indicating that the redness of the cap probably shouldn t be relied upon   In the rare circumstance that a similar shade of silver is behind the bottle  even after the abstraction of the plastic  or the bottle is somehow filled with the same shade of silver liquid  we can fall back on what we can roughly estimate as being the shape of the silver - which as I mentioned  is circular and follows the shape of the can  But even though I lack any certain knowledge in image processing  that sounds slow  Better yet  why not deduce this by for once checking around the sides of the logo to ensure there is nothing of the same silver colour there  Ah  but what if there s the same shade of silver behind a can  Then  we do indeed have to pay more attention to shapes  looking at the top and bottom of the can again    Depending on how flawless this all needs to be  it could be very slow  but I guess my basic concept is to check the easiest and closest things first  Go by colour differences around the already matched shape  which seems the most trivial part of this anyway  before going to the effort of working out the shape of the other elements  To list it  it goes    Find the main attraction  red logo background  and possibly the logo itself for orientation  though in case the can is turned away  you need to concentrate on the red alone  Verify the shape and orientation  yet again via the very distinctive redness Check colours around the shape  since it s quick and painless  Finally  if needed  verify the shape of those colours around the main attraction for the right roundness    In the event you can t do this  it probably means the top and bottom of the can are covered  and the only possible things that a human could have used to reliably make a distinction between the can and the bottle is the occlusion and reflection of the can  which would be a much harder battle to process  However  to go even further  you could follow the angle of the can bottle to check for more bottle-like traits  using the semi-transparent scanning techniques mentioned in the other answers   Interesting additional nightmares might include a can conveniently sitting behind the bottle at such a distance that the metal of it just so happens to show above and below the label  which would still fail as long as you re scanning along the entire length of the red label - which is actually more of a problem because you re not detecting a can where you could have  as opposed to considering that you re actually detecting a bottle  including the can by accident  The glass is half empty  in that case     As a disclaimer  I have no experience in nor have ever thought about image processing outside of this question  but it is so interesting that it got me thinking pretty deeply about it  and after reading all the other answers  I consider this to possibly be the easiest and most efficient way to get it done  Personally  I m just glad I don t actually have to think about programming this   EDIT   Additionally  look at this drawing I did in MS Paint    It s absolutely awful and quite incomplete  but based on the shape and colours alone  you can guess what it s probably going to be  In essence  these are the only things that one needs to bother scanning for  When you look at that very distinctive shape and combination of colours so close  what else could it possibly be  The bit I didn t paint  the white background  should be considered  anything inconsistent   If it had a transparent background  it could go over almost any other image and you could still see it

[c++] Image Processing: Algorithm Improvement for 'Coca-Cola Can' Recognition

Examples related to c++

Examples related to algorithm

Examples related to image-processing

Examples related to opencv