The DOI matches its Crossref record; 47 of the 48 references it cites were found in a registry or an index, and 1 are not checked yet; 4 of the 39 sources its claims cite were checked for support.
Deep Perceptual Hash Based on Hash Center for Image Copyright Protection
Open accessAbstract
At present, most of the perceptual hash methods for image copyright protection rely on manually designed feature extraction and mapping, whose detection accuracy is insufficient. Some schemes based on deep learning are designed to consider a limited variety of content retention operations, which are not enough to deal with the increasingly severe situation of image copyright protection. In response to this situation, a novel Convolutional Neural Network (CNN)-based perceptual image hashing scheme is introduced in this paper. In this scheme, the training images are classified according to their original images and a Hadamard matrix is used to generate a hash center for each class in Hamming space. Then the Convolution Neural Network learns the feature extraction process of the image automatically, constrained by central quantization and distinct quantization, so that the hash code of each image converges to the hash center of their class, and generates the final hash sequence. The proposed scheme can successfully strike a balance between perceived robustness and discrimination capacity. Based on the test results on large-scale test sets,$F_{1}$scores, equal error rate (EER) and receiver operating characteristic (ROC) curves demonstrate the superiority of our scheme compared with some state-of-the-art schemes.
Verified 2026-10-10 · 1 reference not checked yet
Overview
CoChat Summary
TL;DR
The paper proposes a CNN-based perceptual image hashing scheme that treats transformed images from the same original as one class and uses pre-generated hash centers in Hamming space to learn compact hash codes for copyright protection. It reports better F1, EER, and ROC performance than several state-of-the-art schemes while requiring only 1,000 original training images.
Key findings
- The proposed scheme achieved an F1 score of 0.9761 and EER of 0.0331 at 64 bits, compared with DSH (0.9463, 0.0609), CNMC (0.9450, 0.0772), IFNN (0.9101, 0.0894), RP-IVD (0.8367, 0.2014), and DBDH (0.8185, 0.1645).
- It used only 1,000 original images for training and generated 84,000 transformed training images, whereas the paper states some prior methods require more than 200,000 original images.
- On 800 different NUS-WIDE images, pairwise hash distances had a mean of 20.21 and standard deviation of 3.22, while transformed-image distances were mostly below 10 and only a few exceeded 14.06.
- In stolen-image retrieval, when False Positive Ratio was 0.01, the scheme still detected more than 80% of stolen images; with 99,900 and 999,900 irrelevant images, recall decreased to 0.7256 and 0.6885, respectively.
Methods
- The scheme classifies images by original source: images transformed from the same original image form one class, and a Hadamard matrix generates a hash center for each class in Hamming space.
- ResNet50 extracts features, followed by a hash layer of three fully connected layers and ReLU; training is constrained by central quantization, distinct quantization, and a quantization loss.
- Experiments used a newly built high-resolution copyright-protection dataset, USC-SIPI standard images, NUS-WIDE, and ImageNet; comparisons were made at 64 bits using F1, EER, ROC curves, execution time, and retrieval recall.
Limitations
- The paper states that the quantization loss \(L_Q\) may have a negative effect in some cases and needs to be selected case-by-case; the ablation shows \(L_C\) is decisive and \(L_D\) significantly helps.
- The paper states that recall decreases as the number of irrelevant images grows, with recall falling to 0.7256 and 0.6885 for 99,900 and 999,900 irrelevant images, indicating potential for further improvement.
- The paper notes that Hadamard-matrix hash centers require the hash length to be a power of 2, so 24- and 48-bit hash sequences cannot be generated, though 16, 32, and 64 bits are sufficient.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures







![FIGURE 8 ROC curves of our scheme, RP-IVD [20], DSH [29], IFNN [33], CNMC [34] and DBDH [44].](https://figures.cochat.ai/paper-figures/0a223910a5f34796ab715ab52341d45e/deep-perceptual-hash-based-on-hash-center-for-image-copyright-protection-figure-8.png)




Verification
Claims the paper makes
3 Valid · 2 Not valid · 16 Not checked · 56 Own results and interpretations
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“and an image that has been embezzled before the watermark was embedded could not be detected.”
What the sources say
- Could not verifycites [1] [1] J.-E. Lee, Y.-H. Seo, and D.-W. Kim, ''Convolutional neural networkbased digital image watermarking adaptive to the resolution of image and watermark,'' Appl. Sci. , vol. 10, no. 19, p. 6854, Sep. 2020. - sourceExperimental — accuracy not yet measured
The Introduction describes embedding a watermark before storing or distributing an image, then extracting it to claim ownership. It does not address whether theft that occurred before embedding could be detected; searches for pre-embedding theft and detection found no passage making that claim. (LLM, conf=0.68)
- Not checked — no full text of the sourcecites [2] [2] A. Kumar, ''A review on implementation of digital image watermarking techniques using LSB and DWT,'' Information and Communication Technology for Sustainable Development . Berlin, Germany: Springer, 2020, pp. 595 602. - source
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“However, the embedding of digital watermarking is usually accompanied by some degree of image quality degradation”
What the sources say
- Supportedcites [1] [1] J.-E. Lee, Y.-H. Seo, and D.-W. Kim, ''Convolutional neural networkbased digital image watermarking adaptive to the resolution of image and watermark,'' Appl. Sci. , vol. 10, no. 19, p. 6854, Sep. 2020. - source
Section 4.4.1 compares watermarked images with the original host images and reports finite PSNR values and visible differences when magnified. The paper describes the trade-off between watermark invisibility and robustness, supporting the general point that embedding can degrade image quality, even though the degradation in its proposed method is difficult to see. (LLM, conf=0.88)
- Not checked — no full text of the sourcecites [2] [2] A. Kumar, ''A review on implementation of digital image watermarking techniques using LSB and DWT,'' Information and Communication Technology for Sustainable Development . Berlin, Germany: Springer, 2020, pp. 595 602. - source
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“and provide some images after malicious operations and degradation common in infringement for training and testing.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Rests on an assumption not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The addition of LQ may have a negative effect in some cases.”
hedgedstates a possible cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Based on the test results on large-scale test sets, F 1 scores, equal error rate (EER) and receiver operating characteristic (ROC) curves demonstrate the superiority of our scheme compared with some state-of-the-art schemes.”
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
32 more claims
Create a free account to see all 77 claims and every source verdict.
- It may state two separate claims as one. 30
- Probably not a substantial claim. 16
- The quote only partly states this claim. 3
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ H. APPLICATION IN COPYRIGHT PROTECTION · p. 10
“| Recall Ratio | False Positive Ratio=0.1 | 0.9201 | 0.8017 | 0.7965 | 0.1895 | 0.2015 | 0.3146 |”
§ H. APPLICATION IN COPYRIGHT PROTECTION · p. 10
“| Recall Ratio | False Positive Ratio=0.01 | 0.8674 | 0.3751 | 0.6683 | 0.6437 | 0.1023 | 0.1922 |”
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1×
“We adopt ResNet50 [39] to extract image features.”
MethodImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 2×
“In this round of test, we randomly selected 1, 200 original images from the ImageNet [40] image dataset, transformed the images according to the transformation operations in Table 2 to generate a total of 1 , 200 × 200 = 240 , 000 test images, and calculated the distance D between the perceptual…”
BackgroundMethodMicrosoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460 · cited 1×
“Previous studies have generally transformed and generated training sets on the basis of general image datasets such as ImageNet [40], NUS WIDE [41] and MS COCO [42].”
BackgroundDeep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.03385 · Cited by 4,196
NUS-WIDE: a real-world web image database from National University of Singapore
Chua, T.-S., Tang, J., Hong, R., Li, H., Luo, Z., & Zheng, Y. (2009) · Proceedings of the ACM International Conference on Image and Video Retrieval · doi:10.1145/1646396.1646452 · Cited by 3,025 · cited 2×
“Previous studies have generally transformed and generated training sets on the basis of general image datasets such as ImageNet [40], NUS WIDE [41] and MS COCO [42].”
BackgroundMethodImageNet: A large-scale hierarchical image database
Deng, J. H., Dong, W. F., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvprw.2009.5206848 · Cited by 2,714
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Bourdev, L., Girshick, R., Hays, J., Perona, P., Ramanan, D., Zitnick, C. L., & Dollár, P. (2014) · arXiv · doi:10.48550/arxiv.1405.0312 · Cited by 2,579 · cited 1×
“Previous studies have generally transformed and generated training sets on the basis of general image datasets such as ImageNet [40], NUS WIDE [41] and MS COCO [42].”
BackgroundSupervised Hashing for Image Retrieval via Image Representation Learning
Xia, R., Pan, Y., Lai, H., Liu, C., & Yan, S. (2014) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v28i1.8952 · Cited by 952 · cited 1×
“Since the CNNH [28] method was proposed in 2014, more and more image hash related research began to turn to Convolutional Neural Network.”
MethodDeep Supervised Hashing for Fast Image Retrieval
Liu, H., Wang, R., Shan, S., & Chen, X. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.227 · Cited by 933
HashNet: Deep Learning to Hash by Continuation
Cao, Z., Long, M., Wang, J., & Yu, P. S. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.598 · Cited by 777 · cited 1×
“The representative ones are DSH [29], DPSH [30], HashNet [31], ADSH [32] etc.”
BackgroundFeature Learning based Deep Supervised Hashing with Pairwise LabelsPreprint
Li, W.-J., Wang, S., & Kang, W.-C. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1511.03855 · Cited by 741
Deep Hashing Network for Efficient Similarity Retrieval
Zhu, H., Long, M., Wang, J., & Cao, Y. (2016) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v30i1.10235 · Cited by 664 · cited 2×
“In order to refine the generated hash code, we refer to the method in DNH [38] to introduce a bi-modal Laplacian prior as the quantization loss for quantization: LQ = ∑ m i6=j(||| 2hi − 1 | −1 ||1), where 1 ∈ R K is an all-one vector.”
MethodRadisic, A. (2025) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.17117444 · Cited by 462
A robust content based digital signature for image authentication
Schneider, M., & Chang, S.-F. (2002) · Proceedings of 3rd IEEE International Conference on Image Processing · doi:10.1109/icip.1996.560425 · Cited by 421 · cited 1×
“The image hashing method designed by Schneider and Chang using histogram features is the earliest research in this field [18].”
MethodCentral Similarity Quantization for Efficient Image and Video Retrieval
Yuan, L., Wang, T., Zhang, X., Tay, F. E., Jie, Z., Liu, W., & Feng, J. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00315 · Cited by 413 · cited 3×
“We refer to the method in [36] to maximize the likelihood probability to obtain the log maximum a posteriori (MAP) estimate of the hash sequenceH = [h1, · · · , hN ]:”
BackgroundMethodRobust hash functions for digital watermarking
Fridrich, J., & Goljan, M. (2002) · Proceedings International Conference on Information Technology: Coding and Computing (Cat. No.PR00540) · doi:10.1109/itcc.2000.844203 · Cited by 336 · cited 1×
“Fridrich et al. replaced the DCT modes with the low-frequency DC-free random smoothing patterns generated from a secret key, and then quantized it through the threshold to generate a specific length of image hash sequence [7].”
BackgroundAsymmetric Deep Supervised Hashing
Jiang, Q.-Y., & Li, W.-J. (2017) · AAAI Conference on Artificial Intelligence · doi:10.48550/arxiv.1707.08325 · Cited by 289
Asymmetric Deep Supervised Hashing
Jiang, Q.-Y., & Li, W.-J. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11814 · Cited by 266 · cited 1×
“The representative ones are DSH [29], DPSH [30], HashNet [31], ADSH [32] etc.”
BackgroundRobust Image Hashing With Ring Partition and Invariant Vector Distance
Tang, Z., Zhang, X., Li, X., & Zhang, S. (2016) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2015.2485163 · Cited by 247 · cited 5×
“ROC curves of our scheme, RP-IVD [20], DSH [29], IFNN [33], CNMC [34] and DBDH [44]. faster or similar generation speed.”
BackgroundMethodInfluential(2010) · Cited by 230
Content-based image copy detection
Kim, C. (2003) · Signal Processing Image Communication · doi:10.1016/s0923-5965(02)00130-3 · Cited by 165 · cited 1×
“Kim [6] proposed an image copy detection schemes by dividing the image into blocks and combining discrete cosine transform (DCT).”
MethodRobust Image Hashing with Tensor Decomposition
Tang, Z., Chen, L., Zhang, X., & Zhang, S. (2019) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2018.2837745 · Cited by 134 · cited 1×
“In 2019 [22], they designed a novel image hash by constructing a stable third-order tensor from the normalized image and then decomposing the third-order tensor using a tensor decomposition algorithm known as Tuck decomposition.”
MethodComparing the Performance of L*A*B* and HSV Color Spaces with Respect to Color Image SegmentationPreprint
Bora, D. J., Gupta, A. K., & Khan, F. A. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1506.01472 · Cited by 133 · cited 1×
“Methods based on image visual features usually rely on image features such as SIFT [13], HOG [14], GIST [15], LBP [16], HSV [17], etc.”
MethodRobust image hashing using ring-based entropies
Tang, Z., Zhang, X., Huang, L., & Dai, Y. (2013) · Signal Processing · doi:10.1016/j.sigpro.2013.01.008 · Cited by 86 · cited 1×
“In 2013 [19], they proposed to use ring based entropy to build image hashes to resist rotation attacks.”
BackgroundDeep balanced discrete hashing for image retrieval
Zheng, X., Zhang, Y., & Lu, X. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.04.037 · Cited by 83
Robust image hashing using non-uniform sampling in discrete Fourier domain
Qin, C., Chang, C.-C., & Tsou, P.-L. (2013) · Digital Signal Processing · doi:10.1016/j.dsp.2012.11.002 · Cited by 81 · cited 1×
“In [8], robust features are extracted from secondary images after DFT transformation.”
MethodRobust image hashing with multidimensional scaling
Tang, Z., Huang, Z., Zhang, X., & Lao, H. (2017) · Signal Processing · doi:10.1016/j.sigpro.2017.02.008 · Cited by 80 · cited 1×
“In 2017 [21], they used the rotation invariant feature matrix, An image hashing scheme based on multidimensional scaling (MDS) is proposed.”
MethodQin, C., Liu, E., Feng, G., & Zhang, X. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3047142 · Cited by 78 · cited 6×
“ROC curves of our scheme, RP-IVD [20], DSH [29], IFNN [33], CNMC [34] and DBDH [44]. faster or similar generation speed.”
BackgroundMethodInfluentialAcharya, U. R., Fujita, H., Bhat, S., Raghavendra, U., Gudigar, A., Molinari, F., Vijayananthan, A., & Ng, K. H. (2015) · Information Fusion · doi:10.1016/j.inffus.2015.09.006 · Cited by 77 · cited 1×
“Methods based on image visual features usually rely on image features such as SIFT [13], HOG [14], GIST [15], LBP [16], HSV [17], etc.”
MethodEfficient Hashing Method Using 2D-2D PCA for Image Copy Detection
Liang, X., Tang, Z., Huang, Z., Zhang, X., & Zhang, S. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3131188 · Cited by 77 · cited 1×
“Liang et al. found translation invariance, a new property of 2D-2D (bidirectional two-dimensional) PCA (principal component analysis), then designed an efficient hashing method for image copy detection combined with 2D-2D PCA and PCT (Polar Coordinate Transformation) [27].”
MethodLee, J.-E., Seo, Y.-H., & Kim, D.-W. (2020) · Applied Sciences · doi:10.3390/app10196854 · Cited by 76 · cited 1×
“In the past decades, people have relied on various types of digital watermarking [1], [2] to mark protected images and detect suspected infringing images.”
BackgroundAnalysis of various image feature extraction methods against noisy image: SIFT, SURF and HOG
Routray, S., Ray, A. K., & Mishra, C. (2017) · International Conference on Electrical, Computer and Communication Technologies (ICECCT) · doi:10.1109/icecct.2017.8117846 · Cited by 67 · cited 1×
“Methods based on image visual features usually rely on image features such as SIFT [13], HOG [14], GIST [15], LBP [16], HSV [17], etc.”
MethodDeep Supervised Hashing for Fast Image Retrieval
Liu, H., Wang, R., Shan, S., & Chen, X. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01174-4 · Cited by 67
Robust image hashing via DCT and LLE
Tang, Z., Lao, H., Zhang, X., & Liu, K. (2016) · Computers & Security · doi:10.1016/j.cose.2016.07.006 · Cited by 58 · cited 1×
“Tang et al. adopted DCT to extract a stable characteristic matrix, then reduced data of feature matrix with local linear embedding (LLE) [11].”
BackgroundHuang, Z., & Liu, S. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3027001 · Cited by 56 · cited 1×
“They further obtained Normalized histogram and local frequency coefficient features from the maximum gradient magnitude, which improved the processing ability of screen content images (SCIs) [25].”
Background(2010) · doi:10.1117/12.876617 · Cited by 56
A key-dependent secure image hashing scheme by using Radon transform
Ou, Y. D., & Rhee, K.-H. (2009) · International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) · doi:10.1109/ispacs.2009.5383770 · Cited by 54 · cited 1×
“D Discrete Cosine Transform (DCT) and Radon transform [10].”
MethodEfficient Image Hashing with Geometric Invariant Vector Distance for Copy Detection
Liu, S., & Huang, Z. (2019) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3355394 · Cited by 53 · cited 1×
“…the preprocessed image into multiple overlapping blocks, obtained two characteristic matrices through invariant moments and low-frequency discrete cosine transform coef-ficients, and quantized the invariant distance between vectors in the characteristic matrix to form a compact hash code [12].”
BackgroundRobust and Secure Image Fingerprinting Learned by Neural Network
Li, Y., Wang, D., & Tang, L. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2019.2890966 · Cited by 52 · cited 6×
“This method [33] takes the lead in applying deep learning to the field of Perceptual Hashing, but it is not an end-to-end method and the combination of multiple codecs are still manually designed.”
BackgroundMethodInfluentialA Review on Implementation of Digital Image Watermarking Techniques Using LSB and DWT
Kumar, A. (2020) · Advances in Intelligent Systems and Computing · doi:10.1007/978-981-13-7166-0_59 · Cited by 49 · cited 1×
“In the past decades, people have relied on various types of digital watermarking [1], [2] to mark protected images and detect suspected infringing images.”
BackgroundLocal Binary Pattern (LBP) with application to variant object detection: A survey and method
Karis, M. S., Razif, N. R. A., Ali, N. M., Rosli, M. A., Aras, M. S. M., & Ghazaly, M. M. (2016) · IEEE International Colloquium on Signal Processing & Its Applications (CSPA) · doi:10.1109/cspa.2016.7515835 · Cited by 48 · cited 1×
“Methods based on image visual features usually rely on image features such as SIFT [13], HOG [14], GIST [15], LBP [16], HSV [17], etc.”
MethodPerceptual Image Hashing Based on Weber Local Binary Pattern and Color Angle Representation
Qin, C., Hu, Y., Yao, H., Duan, X., & Gao, L. (2019) · IEEE Access · doi:10.1109/access.2019.2908029 · Cited by 48 · cited 1×
“In [24], Chuan used the combination of local texture and color angle features to generate image hash.”
MethodSIFT Matching with CNN Evidences for Particular Object Retrieval
Zhang, G., Zeng, Z., Zhang, S., Zhang, Y., & Wu, W. (2017) · Neurocomputing · doi:10.1016/j.neucom.2017.01.081 · Cited by 39 · cited 1×
“Methods based on image visual features usually rely on image features such as SIFT [13], HOG [14], GIST [15], LBP [16], HSV [17], etc.”
MethodRobust Image Hashing With Isomap and Saliency Map for Copy Detection
Liang, X., Tang, Z., Wu, J., Li, Z., & Zhang, X. (2023) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3139217 · Cited by 32 · cited 1×
“In 2021 [23], They used frequency Tuned (FT) method to determine saliency map for hash operation, and applied Isometric Mapping (isomap) to the field of image hashing for the first time.”
MethodSecure and Traceable Copyright Management System Based on Blockchain
Wang, P., Li, Y., Li, F., Dong, X., & Chen, P. (2019) · IEEE International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc47050.2019.9064101 · Cited by 29 · cited 1×
“Some copyright protection schemes [3], [4] use cryptographic hash functions such as SHA-1, SHA-256 and MD5 to generate image hash codes.”
MethodLearning to Hash for Personalized Image Authentication
Su, Z., Yao, L., Mei, J., Zhou, L., & Li, W. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3002146 · Cited by 27 · cited 1×
“Su et al. Combined distance metric learning technology and supervised quantization strategy into the image authentication process, and proposed a personalized authentication framework [26].”
BackgroundAndi, Purba, R., & Yunis, R. (2019) · International Conference on Informatics and Computing (ICIC) · doi:10.1109/icic47613.2019.8985920 · Cited by 26 · cited 1×
“Some copyright protection schemes [3], [4] use cryptographic hash functions such as SHA-1, SHA-256 and MD5 to generate image hash codes.”
MethodPatil, V. (2019) · Journal of Computers · doi:10.17706/jcp.14.3.210-222 · Cited by 14 · cited 1×
“Patil and Sarode applied the center symmetric local binary pattern (CSLBP) to the LL band of Discrete Wavelet Transform (DWT) to achieve compact length hashing [9].”
Background
Research works printed but not linked: 3
- weisstein, 2002, Hadamard matrix
- liu, 2020, An effective deep hash image copy detection algorithm, Applied Soft Computing, 3, 213
- 2016, The usc-sipi image database
Cited by
Perceptual Authentication Hashing for Digital Images With Contrastive Unsupervised Learning
Gao, G., Qin, C., Fang, Y., & Zhou, Y. (2023) · IEEE Multimedia · doi:10.1109/mmul.2023.3280669 · Cited by 14
TMCIH: Perceptual Robust Image Hashing with Transformer-based Multi-layer Constraints
Fang, Y., Zhou, Y., Li, X., Kong, P., & Qin, C. (2023) · ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec) · doi:10.1145/3577163.3595113 · Cited by 14
“[13] proposed a CNN scheme based on hash center, which treated the original image and perceptually identical images as one category and converged them to their hash centers.”
A Survey of Perceptual Hashing for Multimedia
Zhou, Y., Li, X., Xiong, C., Yao, H., & Qin, C. (2025) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3727880 · Cited by 18
“Sun 𝑒𝑡 𝑎𝑙 .”
Perceptual authentication hashing for digital images based on multi-domain feature fusion
Cao, F., Yao, S., Zhou, Y., Yao, H., & Qin, C. (2024) · Signal Processing · doi:10.1016/j.sigpro.2024.109576 · Cited by 8
Xu, J., Zhang, J., & Wang, J. (2025) · Journal of Theoretical and Applied Electronic Commerce Research · doi:10.3390/jtaer20020076 · Cited by 9
“The third part improves the traditional perceptual hash algorithm for the problems of missing image details during feature capture and low fetch accuracy in determining similarity during image infringement detection [8].”
Metadata
- Authors
- Sun, X., & Zhou, J.
- Affiliations
- Shanghai University (CN)
- Venue
- IEEE Access
- Published
- 2022
- Volume
- 10
- Pages
- 120551-120562
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W4312538097
- Cited by
- 20
- References
- 45
- Citations by year
- 2026: 3 · 2025: 5 · 2024: 5 · 2023: 7
- FWCI
- 1.39
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- IEEE Access (publishedVersion, cc-by-nc-nd)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Steganography and Watermarking Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image Retrieval and Classification Techniques
- Keywords
- perceptual image hashing, image copyright protection, Hadamard matrices, Hamming space, convolutional neural networks
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.