Compact Descriptors for Visual Search for Money Recognition

Undergraduate thesis Compact Descriptors for Visual Search for Money Recognition Student: Pham Tran Huong Giang Supervisor: Dr Le Thanh Ha Major: Computer Science – Faculty: Information Technology University of Engineering and Technology Outline  Introduction  Compact Descriptors for Visual Search  Approach  Conclusion and future work Introduction  Compact Descriptors for Visual Search  International standard for visual search systems  Formed and developed by MPEG  Will be used in a great number of visual search applications Introduction  Money recognition: a practical application for businessman and tourists It is hard to recognize local money in the first use when they come to a foreign country  Motivated application: money recognition for smartphone Study CDVS and evaluate the ability of using CDVS to solve money recognition problem Compact descriptor for visual search  Extracting compact descriptor  Retrieval  Pairwise matching Compact Descriptors for Visual Search  Extracting compact descriptor Detect key points Select features Aggregate global descriptor Encode feature’s coordinate Compact descriptor Describe Compress local descriptors Local descriptors Compact Descriptors for Visual Search  Retrieval Query image Extract compact descriptor Compare global descriptors Descriptor database Top match list Compact Descriptors for Visual Search  Pairwise matching Referent images Query image Extract descriptor Match local descriptors in compressed domain Check the geometric consistency Matching Homography Extract descriptor Approach  To evaluate:  Collect a set of images of money as training set  Build a small program to recognize money  Test by another set of images (test set) Approach  Collecting training dataset  Consists of the money of countries: Vietnam, Laos, Cambodia, Japan, Thailand, Singapore  Good condition of displayed money  Without background  Uniform distributed light  238 images: 160 images of banknote and 78 images of coin 10 Approach  Building money recognition program Input image Find and extract circle 11 Extract descriptor Extract descriptor Match with 160 descriptors of banknote Match with 78 descriptors of coin Evaluate by compare with threshold one of banknote and one of coin Result How much? Nationality? Approach  Testing and discussion  Test set: subsets 12 100 images of banknotes (same type with images in training dataset, taken by me) 50 downloaded images from the Internet (same type with training images) 60 images of coins (same type with training images, taken by me) 25 distractor images of banknotes 10 distractor images of coins 20 distractor images without money Testing result  For threshold of 100 for banknote and 50 for coin Subset Precision Number of “not found” Number of false results results 13 70/100 30 3/60 54 3 46/50 20/20 20 18/20 18 10/10 10 Discussion  Some successful tests in subset 14 Discussion  Some “not found” tests in subset  Comment: Money in these images is folded or strong line shine through the money 15 Discussion  In subset 2: “not found” cases are caused by the low quality of images  One “false” case because of the similarity between images: 16 Discussion  CDVS treats well with distractor images (subset 4, 5, 6)  Only wrong cases in subset because the distractor images are similar with training images 17 Discussion  For coins, the recognition accuracy is hardly acceptable because of some reasons  Coin is poor in feature  Strongly response to light  Whole-colorized  Small surface 18 Conclusion and future work  Conclusion:  CDVS brings high recognition accuracy for banknote if the image of banknote is taken in not too bad condition  The recognition accuracy for coin is hardly acceptable  Future work:  Collect more data  Build an mobile application for tourists to recognize all kind of money in the world 19 Thank you for watching 20 ... Introduction  Compact Descriptors for Visual Search  Approach  Conclusion and future work Introduction  Compact Descriptors for Visual Search  International standard for visual search systems  Formed... local descriptors Local descriptors Compact Descriptors for Visual Search  Retrieval Query image Extract compact descriptor Compare global descriptors Descriptor database Top match list Compact Descriptors. .. application: money recognition for smartphone Study CDVS and evaluate the ability of using CDVS to solve money recognition problem Compact descriptor for visual search  Extracting compact descriptor