Deep Learning for Tattoo Recognition

Di, Xing; Patel, Vishal M.

doi:10.1007/978-3-319-61657-5_10

Cited by 10 publications

(4 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [5] and [6], authors also propose using a CNN for tattoo detection, also basing their study on the Tatt-C dataset. The proposed model consists in extracting features through fine tuning the AlexNet network and, then, applying a linear Support Vector Machine (SVM) to determine whether an image has tattoos or not.…”

Section: Literature Reviewmentioning

confidence: 99%

A Transfer Learning Approach for the Tattoo Detection Problem

Silva¹,

Lopes²

2021

Anais Do 15. Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

Tattoos are still poorly explored as a biometrics factor for human identification, especially in public security, where tattoos can play an important role for identifying criminals and victims. Tattoos are considered a soft biometrics, since they are not permanent and can change along time, differently from hard biometrics traits (fingerprint, iris, DNA, etc). The identification of tattoos are not simple, since they do not have a definite pattern or location. This fact increases the complexity of developing models to address this problem. In addition, the tattoo identification roadmap is very complex, including several steps and, in each step, specific methods need to be developed. Among the several problems identified in this roadmap, we tacked the identification problem, which is defined as: given an image of a person, determine if there is a tattoo or not. We present a deep learning model based on transfer learning for the tattoo detection problem. We also used data augmentation to improve the diversity of the training sets so as to achieve better classification accuracy. Along the work two new datasets for tattoo detection were created. Several comparative experiments were done to evaluate the diversity of images in the datasets, and the accuracy of the proposed model. Results were very promising, achieving an accuracy of 95.1% in the test set, and a F1-score of 0.79 in an external dataset. Overall, results were satisfactory, given the complexity of the problem. Future work will focus on expanding the datasets created and addressing the other problems of the tattoo roadmap.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

A Transfer Learning Approach for the Tattoo Detection Problem

Silva¹,

Lopes²

2021

Anais Do 15. Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

show abstract

“…Identity verification is a broad area with many applications and proposed solutions (see [29], [15], [14], [16]). With the rapid advances made over the last decade in the capabilities of deep neural networks (DNNs), it had become possible to identify people with a very high level of confidence simply by comparing pairs of images and deciding whether they represent the same person or not, even when the two images differ in age, pose, facial expression, hairstyle, and lighting.…”

Section: Introductionmentioning

confidence: 99%

Facial Misrecognition Systems: Simple Weight Manipulations Force DNNs to Err Only on Specific Persons

Zehavi¹,

Shamir²

2023

Preprint

View full text Add to dashboard Cite

In this paper we describe how to plant novel types of backdoors in any facial recognition model based on the popular architecture of deep Siamese neural networks, by mathematically changing a small fraction of its weights (i.e., without using any additional training or optimization). These backdoors force the system to err only on specific persons which are preselected by the attacker. For example, we show how such a backdoored system can take any two images of a particular person and decide that they represent different persons (an anonymity attack), or take any two images of a particular pair of persons and decide that they represent the same person (a confusion attack), with almost no effect on the correctness of its decisions for other persons. Uniquely, we show that multiple backdoors can be independently installed by multiple attackers who may not be aware of each other's existence with almost no interference.We have experimentally verified the attacks on a FaceNet-based facial recognition system, which achieves SOTA accuracy on the standard LFW dataset of 99.35%. When we tried to individually anonymize ten celebrities, the network failed to recognize two of their images as being the same person in 96.97% to 98.29% of the time. When we tried to confuse between the extremely different looking Morgan Freeman and Scarlett Johansson, for example, their images were declared to be the same person in 91.51% of the time. For each type of backdoor, we sequentially installed multiple backdoors with minimal effect on the performance of each one (for example, anonymizing all ten celebrities on the same model reduced the success rate for each celebrity by no more than 0.91%). In all of our experiments, the benign accuracy of the network on other persons was degraded by no more than 0.48% (and in most cases, it remained above 99.30%).

show abstract

“…The first trend exploits cryptographic algorithms to keep integrity of original information by exploiting digital signatures [8], [9]. The second trend refers to the use of tattoo which protects the copyright of information by inserting a signature in the original information [10]. This signature is not directly visible without appropriate technique.…”

Section: Introductionmentioning

confidence: 99%

An Efficient DCT-SVD Steganographic Approach Applied to JPEG Images

Tchakounté¹,

Kamdem²,

Kamgang³

et al. 2020

EAI Endorsed Transactions on Industrial Networks and Intelligen

View full text Add to dashboard Cite

To prove the origin of images in social media, this work proposes an efficient JPEG image steganography approach. After structuring the cover image into blocks of 8*8 pixels, Discrete Cosine Transform is applied to each block of pixels. The latter are quantified using a quantization table and a matrix of DC coefficients from quantized blocks of pixels, is obtained. Singular Value Decomposition is applied to the previous matrix and the secret message is inserted within singular vectors. For extraction purposes, previous transformations are followed reversely. An experimentation is made on seven images and results show that the proposed system outperforms similar studies in three aspects (i) it preserves stego image quality with PSNR of stego images varying between 38 and 54 (ii) it is able to insert a secret message of 257600 bits with the capacity of 4 bits per pixel (iii) it is robust and resistant to attacks such as histogram analysis and chi-square test.

show abstract

Deep Learning for Tattoo Recognition

Cited by 10 publications

References 28 publications

A Transfer Learning Approach for the Tattoo Detection Problem

A Transfer Learning Approach for the Tattoo Detection Problem

Facial Misrecognition Systems: Simple Weight Manipulations Force DNNs to Err Only on Specific Persons

An Efficient DCT-SVD Steganographic Approach Applied to JPEG Images

Contact Info

Product

Resources

About