Efficient Image Copy Detection Using Multiscale Fingerprints

Hefei Ling,Hongrui Cheng,Qingzhen Ma,Fuhao Zou,Weiqi Yan

DOI: https://doi.org/10.1109/mmul.2011.75

IF: 3.4911

2012-01-01

IEEE Multimedia

Abstract:A multiscale scale-invariant feature transform (SIFT) descriptor can help improve our ability to discriminate between images when using copy detection to identify illegal image copies.

What problem does this paper attempt to address?

Detecting Image Tampering Using Feature Fusion

Pin Zhang,Xiangwei Kong

DOI: https://doi.org/10.1109/ares.2009.150

2009-01-01

Abstract:Along with the development of sophisticated image processing software, it is getting easier forging a digital image but harder to detect it. It is already a problem for us to distinguish tampered photos from authentic ones. In this paper, we propose an approach based on feature fusion to detect digital image tampering. First, we extract the feature statistics that can represent the property of a camera from the images taken by that camera. These feature statistics are used for training a one-class classifier in order to get the feature pattern of the given camera. Then, we do sliding segmentation to testing images. Finally, feature statistics extracted from image blocks are fed into the trained one-class classifier to match the feature pattern of the given camera. The images with low percentage of matched blocks are classified as tampered ones. Our method could achieve a high accuracy in detecting the tampered images that undergone post-processing such as JPEG compression, re-sampling and retouching.
Fine-search for image copy detection based on local affine-invariant descriptor and spatial dependent matching

Hefei Ling,Liyun Wang,Fuhao Zou,Weiqi Yan

DOI: https://doi.org/10.1007/s11042-009-0439-9

IF: 2.577

2010-01-01

Multimedia Tools and Applications

Abstract:Copies are somehow a subset of near-duplicates, but the approaches extensively employed in near-duplicate retrieval only obtain rough and imprecise query results. Therefore a fine-search scheme is proposed to refine these rough results and attempt to completely detect the real copies accurately. This scheme first employs a local affine-invariant descriptor based on polar-mapping and discrete Fourier transform. Then a spatial dependent matching method is proposed combining nearest neighbor distance ratio with the spatial relationships among the local features. Experimental results demonstrate that the employed descriptor is more robust, distinctive and suitable for copy detection in comparison with the SIFT descriptor. And the spatial dependent matching is able to improve the recall and precision, and lower the false positives and ambiguities.
High-performance image forgery detection via adaptive SIFT feature extraction for low-contrast or small or smooth copy–move region images

J. S. Sujin,S. Sophia

DOI: https://doi.org/10.1007/s00500-023-08209-6

IF: 3.732

2023-04-15

Soft Computing

Abstract:Due to their robustness against large-scale geometric transformations, keypoint-based detection methods play an important role in revealing copy–move evidence. However, where copy–move forgeries are only involved in low-contrast or small or smooth regions, these approaches do not yield high-performance results because the number of keypoints in these regions is very limited. We recommend a highly efficient copy–move forgery detection algorithm by ADaptive Scale-Invariant Feature Transform (ADSIFT). Initially, by adapting the gamma factor for contrast threshold and rescaling factor values for feature matching and forgery detection, we produce an adequate number of keypoints that occur even in low-contrast or small or smooth regions of noisy or noiseless images. The proposed approach delivers greater performance than the existing methods as compared to precision and efficiency.

computer science, artificial intelligence, interdisciplinary applications
Fogery image blind detection by asymemetric search based on SIFT

Du Zhenlong,Yang Fan,Li Xiaoli,Guo Yanwen,Shen Ganggang

2013-01-01

Journal of Image and Graphics

Abstract:The identification of copy-paste forgery image is to find the suspicious region via pixel-by-pixel or block-by-block match,the computation costs is very heavy.An efficient blind forgery image detection approach based on scale-inva-riant feature transform(SIFT)is proposed in the paper,which employs SIFT keypoints for positioning the initial suspicious forgery region.The asymmetric search is exploited for refining the suspicious region and determining the forgery area.Experiments demonstrate that the proposed algorithm could significantly decrease the number of candidate search block,accurately identify the copy-paste forgery region regardless of the existence of gauss,salt and pepper noise,and the computational cost is reduced by 1～2 orders of magnitude with compare to the conventional methods.
Fast and robust copy-move forgery detection based on scale-space representation

Chun-Su Park,Joon Yeon Choeh

DOI: https://doi.org/10.1007/s11042-017-5248-y

IF: 2.577

2017-10-02

Multimedia Tools and Applications

Abstract:Copy-move forgery (CMF), which copies a part of an image and pastes it into another region, is one of the most common methods for digital image tampering. For CMF detection (CMFD), we propose a fast and robust approach that can handle several geometric transformations including rotation, scaling, sheering, and reflection. In the proposed CMFD design, keypoints and their descriptors are extracted from the image based on the Scale Invariant Feature Transform (SIFT). Then, an improved matching operation that can handle multiple copy-move forgeries is performed to detect matched pairs located in duplicated regions. Next, the geometric transformation between duplicated regions is estimated using a subset of reliable matched pairs which are obtained using the SIFT scale space representation. In our simulation, we present comparative results between the proposed algorithm and state-of-the-art ones with proven performance guarantees.

computer science, information systems, theory & methods,engineering, electrical & electronic, software engineering
Copy-Move Forgery Detection Using Scale Invariant Feature and Reduced Local Binary Pattern Histogram

Jun Young Park,Tae An Kang,Yong Ho Moon,Il Kyu Eom

DOI: https://doi.org/10.3390/sym12040492

2020-03-26

Symmetry

Abstract:Because digitized images are easily replicated or manipulated, copy-move forgery techniques are rendered possible with minimal expertise. Furthermore, it is difficult to verify the authenticity of images. Therefore, numerous efforts have been made to detect copy-move forgeries. In this paper, we present an improved region duplication detection algorithm based on the keypoints. The proposed algorithm utilizes the scale invariant feature transform (SIFT) and the reduced local binary pattern (LBP) histogram. The LBP values with 256 levels are obtained from the local window centered at the keypoint, which are then reduced to 10 levels. For a keypoint, a 138-dimensional is generated to detect copy-move forgery. We test the proposed algorithm on various image datasets and compare the detection accuracy with those of existing methods. The experimental results demonstrate that the performance of the proposed scheme is superior to that of other tested copy-move forgery detection methods. Furthermore, the proposed method exhibits a uniform detection performance for various types of test datasets.
Digital Image Forensics Using Multi-Resolution Histograms

Jin Liu,Hefei Ling,Fuhao Zou,Weiqi Yan,Zhengding Lu

DOI: https://doi.org/10.4018/jdcf.2010100103

2010-01-01

International Journal of Digital Crime and Forensics

Abstract:In this paper, the authors investigate the prospect of using multi-resolution histograms (MRH) in conjunction with digital image forensics, particularly in the detection of two kinds of copy-move manipulations, i.e., cloning and splicing. To the best of the authors’ knowledge, this is the first work that uses the same feature in both cloning and splicing forensics. The experimental results show the simplicity and efficiency of using MRH for the purpose of clone detection and splicing detection.
A Sift-Based Image Fingerprinting Approach Robust to Geometric Transformations

Xinghua Yu,Tiejun Huang

DOI: https://doi.org/10.1109/iscas.2009.5118093

2009-01-01

Abstract:Among approaches in implementing Digital Rights Management, image fingerprinting technique is considered to be one of the most attractive solutions, especially in detecting illegal use of image works. The deficiency of the existing image fingerprinting methods is that they can not deal with geometric transformations, such as aspect ratio changes, rotations, cropping, combining, etc. Aiming at this shortcoming, we propose a SIFT-based image fingerprinting algorithm which is robust to geometric transformations. Firstly, we introduce SIFT-based algorithm to extract features as a unique fingerprint. Secondly, a method based on area ratio invariance of affine transformation is utilized to verify valid matched keypoint pairs between the queried image and the pre-registered image. Finally, by counting the valid matched pairs, we estimate whether the two images are homologous or not. Experimental results demonstrate that the proposed method exhibits an excellent performance when geometric transformation occurs.
Fast SIFT algorithm based on Sobel edge detector

Yang Li,Lingshan Liu,Lianghao Wang,Dongxiao Li,Ming Zhang

DOI: https://doi.org/10.1109/CECNet.2012.6201824

2012-01-01

Abstract:The SIFT (Scale Invariant Feature Transform) algorithm is an approach for extracting distinctive invariant features from images. It is widely used in image matching. Since SIFT detector detects the extreme points through the whole scale space, it often selects keypoints that have no value and reduces the efficiency of the algorithm. This paper proposes a fast SIFT algorithm based on Sobel edge detector. Sobel edge detector is applied to generate an edge group scale space and SIFT detector detects the extreme point under the constraint of the edge group scale space. The experimental results show that the proposed algorithm decreases the redundancy of keypoints and speeds up the implementation while the matching rate between different images maintains at a high level. As the threshold of Sobel detector increases, number of keypoints decreases and matching rate gets higher.
Copy-move forgery detection by improved SIFT K-means algorithm

Kavita Rathi,Parvinder Singh

DOI: https://doi.org/10.1504/ijcvr.2024.139547

2024-07-07

International Journal of Computational Vision and Robotics

Abstract:Copy-move forgery, a type of forgery in which it copies features from the same image, poses the toughest challenge in image forgery detection. Key-point-based CMFD techniques outperform the block-based CMFD techniques. SIFT is the mostly used key-point-based techniques. The present algorithm improves upon the SIFT algorithm with improvements in the various steps of the workflow by adding Laplace of Gaussian and multiplying it by the square of the Gaussian kernel to make it real scale invariant, applying double level filtering at feature extraction, and filtering by using g2NN and K-mean clustering. The results in the form of recall, precision, and F1 measure outperformed the state-of-art key-point-based CMFD techniques over multiple datasets.
An Effective Image Copy-Move Forgery Detection Using Entropy Information

Li Jiang,Zhaowei Lu

2024-04-30

Abstract:Image forensics has become increasingly crucial in our daily lives. Among various types of forgeries, copy-move forgery detection has received considerable attention within the academic community. Keypoint-based algorithms, particularly those based on Scale Invariant Feature Transform, have achieved promising outcomes. However, most of keypoint detection algorithms failed to generate sufficient matches when tampered patches were occurred in smooth areas, leading to insufficient matches. Therefore, this paper introduces entropy images to determine the coordinates and scales of keypoints based on Scale Invariant Feature Transform detector, which make the pre-processing more suitable for solving the above problems. Furthermore, an overlapped entropy level clustering algorithm is developed to mitigate the increased matching complexity caused by the non-ideal distribution of gray values in keypoints. Experimental results demonstrate that our algorithm achieves a good balance between performance and time efficiency.

Computer Vision and Pattern Recognition,Computers and Society,Multimedia
Copy-move forgery detection for image forensics using the superpixel segmentation and the Helmert transformation

Hui-Yu Huang,Ai-Jhen Ciou

DOI: https://doi.org/10.1186/s13640-019-0469-9

IF: 2.399

2019-06-13

EURASIP Journal on Image and Video Processing

Abstract:The increasing popularity of the internet suggests that digital multimedia has become easier to transmit and acquire more rapidly. This also means that this multimedia has become more susceptible to tampering through forgery. One type of forgery, known as copy-move duplication, is a specified type that usually involves image tampering. In this study, a keypoint-based image forensics approach based on a superpixel segmentation algorithm and Helmert transformation has been proposed. The purpose of this approach is to detect copy-move forgery images and to obtain forensic information. The procedure of the proposed approach consists of the following phases. First, we extract the keypoints and their descriptors by using a scale-invariant feature transform (SIFT) algorithm. Then, based on the descriptor, matching pairs will be obtained by calculating the similarity between keypoints. Next, we will group these matching pairs based on spatial distance and geometric constraints via Helmert transformation to obtain the coarse forgery regions. Then, we refine these coarse forgery regions and remove mistakes or isolated areas. Finally, the forgery regions can be localized more precisely. Our proposed approach is a more robust solution for scaling, rotation, and compression forgeries. The experimental results obtained from testing different datasets demonstrate that the proposed method can obtain impressive precision/recall rates in comparison to state-of-the-art methods.

engineering, electrical & electronic,imaging science & photographic technology
An Image Fingerprinting Method Robust to Complicated Image Modifications

Xinghua Yu,Tiejun Huang

DOI: https://doi.org/10.1109/iih-msp.2008.93

2008-01-01

Abstract:Image fingerprinting technique plays an important role in Digital Rights Management and related applications, especially in the detection of the illegal use of image works. Although existing image fingerprinting methods perform well in many cases, they are not robust enough to sophisticated modifications like image embedding and image combining. This paper proposes an image fingerprinting approach which is robust to such modifications. Firstly, we introduce an SIFT-based algorithm to extract image keypoints as the unique fingerprint. Secondly, a technique based on geometry isomorphic relationship is utilized to select valid keypoints in the queried image. Finally, by calculating the matched keypoints pairs between the queried image and the pre-registered image, we can make the decision whether they are homologous. Experimental results show that the proposed method is robust to image embedding and combining, without a high computational complexity.
Copy-Move Image Forgery Detection Using Phase Adaptive Spatio-structured SIFT Algorithm

Raimoni Hansda,Rajashree Nayak,Bunil Kumar Balabantaray,Sonali Samal

DOI: https://doi.org/10.1007/s42979-021-00903-2

2021-11-03

SN Computer Science

Abstract:This paper suggests a hybrid Copy-Move image forgery detection method using phase adaptive Spatio-structured SIFT algorithm which enables to localize forgery regions (FRs) in the presence of combination of intermediate and post-processing attacks. Moreover, the proposed method well localizes varied types and mixed-sized FRs at greater accuracy. Key contributions of the proposed method are: (i) detection of matched key-points via Spatio-structured SIFT (S-SIFT) algorithm; (ii) formation of larger blocks around the matched key-points followed by the division of larger blocks into several non-overlapping blocks; (iii) representation of non-overlapping blocks via feature descriptors based on the histogram of oriented phase congruency (HoPC); (iv) block matching via 2NN matching strategy to localize the final forged areas. S-SIFT algorithm integrates spatial and structural information additionally in the SIFT feature descriptor for detecting sufficient numbers of key-points at relatively smoothed and little structured FRs. Formation of larger blocks around the matched key-points enables to enhance the probability of detecting any sized FRs, whereas non-overlapping division of blocks generates a lesser number of blocks to be matched. Hence, the computational time of the matching process will be reduced. Feature descriptors based on HoPC are robust to noise, invariant to various geometric transforms, insensitive to change in image contrast and non-uniform illumination. Performance assessment of the proposed work in the presence of different attacks is validated both at image-level and pixel-level on Benchmark datasets, such as MICC-F220, GRIP, and CoMoFoD, and found to outperform the state-of-the-art methods.
Image Copy-Move Forgery Detection Based on Fused Features and Density Clustering

Guiwei Fu,Yujin Zhang,Yongqi Wang

DOI: https://doi.org/10.3390/app13137528

2023-01-01

Abstract:Image copy-move forgery is a common simple tampering technique. To address issues such as high time complexity in most copy-move forgery detection algorithms and difficulty detecting forgeries in smooth regions, this paper proposes an image copy-move forgery detection algorithm based on fused features and density clustering. Firstly, the algorithm combines two detection methods, speeded up robust features (SURF) and accelerated KAZE (A-KAZE), to extract descriptive features by setting a low contrast threshold. Then, the density-based spatial clustering of applications with noise (DBSCAN) algorithm removes mismatched pairs and reduces false positives. To improve the accuracy of forgery localization, the algorithm uses the original image and the image transformed by the affine matrix to compare similarities in the same position in order to locate the forged region. The proposed method was tested on two datasets (Ardizzone and CoMoFoD). The experimental results show that the method effectively improved the accuracy of forgery detection in smooth regions, reduced computational complexity, and exhibited strong robustness against post-processing operations such as rotation, scaling, and noise addition.
A two-stage detection method of copy-move forgery based on parallel feature fusion

Wujian Ye,Qingyuan Zeng,Yihang Peng,Yijun Liu,Chin-Chen Chang

DOI: https://doi.org/10.1186/s13638-022-02112-8

2022-04-01

EURASIP Journal on Wireless Communications and Networking

Abstract:Abstract The copy-move forgery refers to the copying and pasting of a region of the original image into the target region of the same image, which represents a typical tampering method with the characteristics of easy tampering and high-quality tampering. The existing single feature-based methods of forgery detection have certain shortcomings, such as high false alarm rate, low robustness, and low detection accuracy. To address these shortcomings, this paper proposes an improved two-stage detection method based on parallel feature fusion and an adaptive threshold generation algorithm. Firstly, the SLIC super-pixels segmentation algorithm is used for image preprocessing, and a similar region extraction algorithm without threshold is employed to obtain the suspected tampering regions with high similarity. Secondly, the parallel fusion feature is obtained based on the SIFT and HU features to express the characteristics of local regions. Then, the corresponding threshold value is generated based on the histogram of oriented gradient (HOG) to describe the texture characteristics of the obtained regions, which acts as a criterion to judge whether a region has been forged or not. The experimental results show that the proposed method outperforms the existing methods, achieving the accuracy of 99.01% and 98.5% on the MICC-F220 and MICC-F2000 datasets respectively. In addition, the proposed method has stronger robustness performance on COMOFOD dataset than the comparison methods.

telecommunications,engineering, electrical & electronic
Advancing Copy-Move Manipulation Detection in Complex Image Scenarios Through Multiscale Detector

Anjali Diwan,Rajesh Mahadeva,Vinay Gupta

DOI: https://doi.org/10.1109/access.2024.3397466

IF: 3.9

2024-05-15

IEEE Access

Abstract:This research presents a new approach for identifying instances of copy-move forgeries in digital images by utilizing the Multiscale Detector a Neural Network-based method, which serves as an image key-point detector and descriptor. The act of copy-move manipulation involves the replication and subsequent insertion of a specific segment of an image, intending to modify the overall content of the image. The approach we utilize leverages the sophisticated functionalities of Multiscale Detector, a framework that combines key-point detection with descriptor extraction, to accurately detect and localize instances of copy-move manipulation. The effectiveness of our approach is assessed on a range of copy-move forgeries, encompassing instances that have undergone post-processing and geometric transformations. The experimental findings illustrate the resilience of our approach in identifying instances of manipulations over a diverse range of textured images and various alteration approaches. Furthermore, our approach demonstrates strong performance even when subjected to supplementary processing procedures such as brightness modification, color reduction, contrast adjustment, and blurring. Our suggested method has greater performance when compared to the existing manipulation detection approach, as demonstrated through a comparative analysis. In addition, the algorithm we have developed has high computing efficiency, allowing for real-time detection of forgeries. The methodology employed in this study, which builds upon the Multiscale Detector framework, offers a highly effective approach to the detection of copy-move manipulations in digital images.

computer science, information systems,telecommunications,engineering, electrical & electronic
Copy-move image forgery detection based on evolving circular domains coverage

Shilin Lu,Xinghong Hu,Chengyou Wang,Lu Chen,Shulu Han,Yuejia Han

DOI: https://doi.org/10.1007/s11042-022-12755-w

IF: 2.577

2022-04-22

Multimedia Tools and Applications

Abstract:Abstract The aim of this paper is to improve the accuracy of copy-move forgery detection (CMFD) in image forensics by proposing a novel scheme and the main contribution is evolving circular domains coverage (ECDC) algorithm. The proposed scheme integrates both block-based and keypoint-based forgery detection methods. Firstly, the speed-up robust feature (SURF) in log-polar space and the scale invariant feature transform (SIFT) are extracted from an entire image. Secondly, generalized 2 nearest neighbor (g2NN) is employed to get massive matched pairs. Then, random sample consensus (RANSAC) algorithm is employed to filter out mismatched pairs, thus allowing rough localization of counterfeit areas. To present these forgery areas more accurately, we propose the efficient and accurate ECDC algorithm to present them. This algorithm can find satisfactory threshold areas by extracting block features from jointly evolving circular domains, which are centered on matched pairs. Finally, morphological operation is applied to refine the detected forgery areas. Experimental results indicate that the proposed CMFD scheme can achieve better detection performance under various attacks compared with other state-of-the-art CMFD schemes.

computer science, information systems, theory & methods,engineering, electrical & electronic, software engineering
Video Copy Detection Using a Soft Cascade of Multimodal Features

Menglin Jiang,Yonghong Tian,Tiejun Huang

DOI: https://doi.org/10.1109/ICME.2012.189

2012-01-01

Abstract:In the video copy detection task, it is widely recognized that none of any single feature can work well for all transformations. Thus more and more approaches adopt a set of complementary features to cope with complex audio-visual transformations. However, most of them utilize individual features separately and the final result is obtained by fusing results of several basic detectors. Often, this will lead to low detection efficiency. Moreover, there are some thresholds or parameters to be elaborately tuned. To address these problems, we propose a soft cascade approach to integrate multiple features for efficient copy detection. In our approach, basic detectors are organized in a cascaded framework, which processes a query video in sequence until one detector asserts it as a copy. To fully exert the complementarity of these detectors, a learning algorithm is proposed to estimate the optimal decision thresholds in the cascade architecture. Excellent performance on the benchmark dataset of TRECVid 2011 CBCD task demonstrates the effectiveness and efficiency of our approach.
Fingerprint Indexing with Minutiae-Aided Fingerprint Multiscale Representation.

Song Wu,Zheng Wang,Zexi Jia,Chuanwei Huang,Hongyan Fei,Jufu Feng

DOI: https://doi.org/10.1007/978-3-031-47634-1_26

2023-01-01

Abstract:The huge fingerprint database limits the speed of large-scale fingerprint identification, which can be alleviated by fingerprint indexing. The previous fixed-length representation used for fingerprint indexing achieves unsatisfactory performance due to no good use of fingerprint domain knowledge. Inspired by some attempts to introduce minutiae information, in this paper, we propose a tailored framework to extract the minutiae-aided fingerprint multiscale representation for fingerprint indexing. We design Minutiae Distribution and Directions Map (MDDM) to encode the information contained in minutiae. Inputting into the network along with the fingerprint images, the MDDM guide the network to pay more attention to the minutiae information. Considering the deep features with high resolution contain more small object information, i.e., the minutiae, we introduce the tailored Multiscale Feature Fusion (MFF) module to better use the semantic information in different layers. Extensive experiments show that the proposed fixed-length representation achieves better indexing performance on all benchmarks.

Efficient Image Copy Detection Using Multiscale Fingerprints

Detecting Image Tampering Using Feature Fusion

Fine-search for image copy detection based on local affine-invariant descriptor and spatial dependent matching

High-performance image forgery detection via adaptive SIFT feature extraction for low-contrast or small or smooth copy–move region images

Fogery image blind detection by asymemetric search based on SIFT

Fast and robust copy-move forgery detection based on scale-space representation

Copy-Move Forgery Detection Using Scale Invariant Feature and Reduced Local Binary Pattern Histogram

Digital Image Forensics Using Multi-Resolution Histograms

A Sift-Based Image Fingerprinting Approach Robust to Geometric Transformations

Fast SIFT algorithm based on Sobel edge detector

Copy-move forgery detection by improved SIFT K-means algorithm

An Effective Image Copy-Move Forgery Detection Using Entropy Information

Copy-move forgery detection for image forensics using the superpixel segmentation and the Helmert transformation

An Image Fingerprinting Method Robust to Complicated Image Modifications

Copy-Move Image Forgery Detection Using Phase Adaptive Spatio-structured SIFT Algorithm

Image Copy-Move Forgery Detection Based on Fused Features and Density Clustering

A two-stage detection method of copy-move forgery based on parallel feature fusion

Advancing Copy-Move Manipulation Detection in Complex Image Scenarios Through Multiscale Detector

Copy-move image forgery detection based on evolving circular domains coverage

Video Copy Detection Using a Soft Cascade of Multimodal Features

Fingerprint Indexing with Minutiae-Aided Fingerprint Multiscale Representation.