Proceedings Article•
Still image copy detection algorithm robust to basic image modifications
01 Sep 2008-Vol. 2, pp 455-458
TL;DR: In this paper, the authors present a method for content based still image replica detection using a compact image signature which depends on image content and is invariant to many widely used image processing techniques, such as lossy compression, resizing, resampling, color enhancements and simple rotations.
Abstract: The paper presents a method for content based still-image replica detection. This method uses a compact image signature which depends on image content and is invariant to many widely used image processing techniques, such as lossy compression, resizing, resampling, color enhancements and simple rotations. The signature is designed to be usable in big image database: it has small size (a few dozen bytes), the extraction is fast and the comparison of image signatures is very fast. More than million of image signatures per second can be compared on a modern PC. Usage of the method within a framework for digital rights infringements detection in the World Wide Web is also discussed.
Citations
More filters
TL;DR: A concentric circle‐based image signature is proposed which makes it possible to detect near‐duplicates rapidly and accurately in large databases and is robust to various modifications.
Abstract: Many applications dealing with image management need a technique for removing duplicate images or for grouping related (near-duplicate) images in a database. This paper proposes a concentric circle-based image signature which makes it possible to detect near-duplicates rapidly and accurately. An image is partitioned by radius and angle levels from the center of the image. Feature values are calculated using the average or variation between the partitioned sub-regions. The feature values distributed in sequence are formed into an image signature by hash generation. The hashing facilitates storage space reduction and fast matching. The performance was evaluated through discriminability and robustness tests. Using these tests, the particularity among the different images and the invariability among the modified images are verified, respectively. In addition, we also measured the discriminability and robustness by the distribution analysis of the hashed bits. The proposed method is robust to various modifications, as shown by its average detection rate of 98.99%. The experimental results showed that the proposed method is suitable for near-duplicate detection in large databases.
5 citations
Proceedings Article•
02 Jun 2017TL;DR: It appears that this trajectory is invariant to many image operations, and the trajectories of original images and distorted copies are highly correlated, which makes the proposed method a good tool for image replica detection.
Abstract: Image replica detection system can be used by the owners of digital multimedia content to protect their rights against unauthorised use of their material. The paper presents a new approach for content-based image replica detection. A concept of singular energy trajectory is introduced and evaluated. It appears that this trajectory is invariant to many image operations. Moreover, the trajectories of original images and distorted copies are highly correlated. These properties make the proposed method a good tool for image replica detection.
4 citations
01 Jan 2009
TL;DR: The chapter presents a method for content based image replica detection, in which a compact image signature is extracted and can be characterized by the following properties: small size, fast extraction and matching, high detection rate for basic image processing techniques.
Abstract: The chapter presents a method for content based image replica detection. In this method a compact image signature is extracted. The signature depends on image content, carries distinctive image information, and is invariant to many widely used image processing techniques, which do not lead to significant loss of information, such as lossy compression, resizing, color enhancements and simple rotations. The detection of unmodified and modified image replicas is performed by matching signatures of query images to signatures of original images. The signature is designed to be usable in big image databases and even in video databases. It can be characterized by the following properties: small size (a few dozen bytes), fast extraction and matching, high detection rate for basic image processing techniques. A few millions of signatures per second can be compared on a modern PC.
1 citations
TL;DR: An image description and matching scheme using synthetic features for a recommendation service where an image is described by synthesized spatial and statistical features and each feature is better than the compared algorithms that use spatial or statistical features.
Abstract: This paper presents an image description and matching scheme using synthetic features for a recommendation service. The recommendation service is an example of smart search because it offers something before a user’s request. In the proposed extraction scheme, an image is described by synthesized spatial and statistical features. The spatial feature is designed to increase the discriminability by reflecting delicate variations. The statistical feature is designed to increase the robustness by absorbing small variations. For extracting spatial features, we partition the image into concentric circles and extract four characteristics using a spatial relation. To extract statistical features, we adapt three transforms into the image and compose a 3D histogram as the final statistical feature. The matching schemes are designed hierarchically using the proposed spatial and statistical features. The result shows that each feature is better than the compared algorithms that use spatial or statistical features. Additionally, if we adapt the proposed whole extraction and matching scheme, the overall performance will become 98.44% in terms of the correct search ratio. Keywords: Image description, image matching, recommendation service, spatial feature, statistical feature.
1 citations
References
More filters
10 Oct 2004
TL;DR: It is shown that, by optimizing layout and access to the index data on disk, the system can efficiently query indices containing millions of keypoints and make approximate similarity queries that only examine a small fraction of the database.
Abstract: We introduce a system for near-duplicate detection and sub-image retrieval. Such a system is useful for finding copyright violations and detecting forged images. We define near-duplicate as images altered with common transformations such as changing contrast, saturation, scaling, cropping, framing, etc. Our system builds a parts-based representation of images using distinctive local descriptors which give high quality matches even under severe transformations. To cope with the large number of features extracted from the images, we employ locality-sensitive hashing to index the local descriptors. This allows us to make approximate similarity queries that only examine a small fraction of the database. Although locality-sensitive hashing has excellent theoretical performance properties, a standard implementation would still be unacceptably slow for this application. We show that, by optimizing layout and access to the index data on disk, we can efficiently query indices containing millions of keypoints. Our system achieves near-perfect accuracy (100% precision at 99.85% recall) on the tests presented in Meng et al. [16], and consistently strong results on our own, significantly more challenging experiments. Query times are interactive even for collections of thousands of images.
432 citations
05 Oct 1998
TL;DR: This paper describes RIME (Replicated IMage dEtector), an alternative approach to watermarking for detecting unauthorized image copying on the Internet and shows that it can detect image copies effectively.
Abstract: This paper describes RIME (Replicated IMage dEtector), an alternative approach to watermarking for detecting unauthorized image copying on the Internet. RIME profiles internet images and stores the feature vectors of the images and their URLs in its repository. When a copy detection request is received, RIME matches the requested image's feature vector with the vectors stored in the repository and returns a list of suspect URLs. RIME characterizes each image using Daubechies' wavelets. The wavelet coefficients are stored as the feature vector. RIME uses a multidimensional extensible hashing scheme to index these high-dimensional feature vectors. Our preliminary result shows that it can detect image copies effectively: It can find the top suspects and copes well with image format conversion, resampling, and requantization.
150 citations
18 Jun 2003
TL;DR: Three schemes are proposed: thresholding, sampling and weighting, and hybrid schemes of these three basic approaches, which substantially outperform DPF in near-replica image recognition.
Abstract: Dynamic Partial Function (DPF), which dynamically selects a subset of features to measure pairwise image similarity, has been shown to be very effective in near-replica image recognition. DPF, however, suffers from the one-size-fits-all problem: it requires that all pairwise similarity measurements must use the same number of features. We propose methods for enhancing DPF's performance by allowing different numbers of features to be selected in a pairwise manner. Through extensive empirical studies, we show that our three schemes: thresholding, sampling and weighting, and hybrid schemes of these three basic approaches, substantially outperform DPF in near-replica image recognition.
47 citations
TL;DR: A system for image replica detection to adapt a system for detecting the replica of a specific reference image and is able to classify test images as replicas of the reference image or as unrelated images.
Abstract: This paper presents a system for image replica detection. The idea behind the proposed approach is to adapt a system for detecting the replica of a specific reference image. The system is then able to classify test images as replicas of the reference image or as unrelated images. More precisely, the test procedure is as follows. A set of features is extracted from a test image, representing texture, colour and grey-level characteristics. These features are then feed into a preprocessing step, which is fine-tuned to the reference image. Finally, the resulting features are entered to a support vector classifier that determines if the test image is a replica of the reference image. Experimental results show the effectiveness of the proposed system. Target applications include search for copyright infringement (e.g. variations of copyrighted images) and known illicit content (e.g. paedophile images known to the police).
12 citations
Proceedings Article•
01 Jan 20074 citations