An official website of the United States government
Abstract— Recent advances show the wide-ranging applications of machine learning for solving multi-disciplinary problems in cancer cell growth detection, modeling cancer growths and treatments, etc. There is growing interests among the faculty and students at Clayton State University to study the applications of machine learning for medical imaging and propose new algorithms based on a recently funded NSF grant proposal in medical imaging, skin cancer detection, and associated smartphone apps and a web-based user-friendly diagnosis interface. We tested many available open-source ML algorithm-based software sets in Python as applied to medical image data processing, and modeling used to predict cancer growths and treatments. We study the use of ML concepts that promote efficient, accurate, secure computation over medical images, identifying and classifying cancer cells, and modeling the cancer cell growths. In this collaborative project with another university, we follow a holistic approach to data analysis leading to more efficient cancer detection based upon both cell analysis and image recognition. Here, we compare ML based software methods and analyze their detection accuracy. In addition, we acquire publicly available data of cancer cell image files and analyze using deep learning algorithms to detect benign and suspicious image samples. We apply the current pattern matching algorithms and study the available data with possible diagnosis of cancer types.
more »
« less
Focal stack based image forgery localization
Image security is becoming an increasingly important issue due to advances in deep learning based image manipulations, such as deep image inpainting and deepfakes. There has been considerable work to date on detecting such image manipulations using improved algorithms, with little attention paid to the possible role that hardware advances may have for improving security. We propose to use a focal stack camera as a novel secure imaging device, to the best of our knowledge, that facilitates localizing modified regions in manipulated images. We show that applying convolutional neural network detection methods to focal stack images achieves significantly better detection accuracy compared to single image based forgery detection. This work demonstrates that focal stack images could be used as a novel secure image file format and opens up a new direction for secure imaging.
more »
« less
- Award ID(s):
- 1838179
- PAR ID:
- 10394999
- Publisher / Repository:
- Optical Society of America
- Date Published:
- Journal Name:
- Applied Optics
- Volume:
- 61
- Issue:
- 14
- ISSN:
- 1559-128X; APOPAI
- Format(s):
- Medium: X Size: Article No. 4030
- Size(s):
- Article No. 4030
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Recent advances show the wide-ranging applications of machine learning for solving multi-disciplinary problems in cancer cell growth detection, modeling cancer growths and treatments, etc. There is growing interests among the faculty and students at Clayton State University to study the applications of machine learning for medical imaging and propose new algorithms based on a recently funded NSF grant proposal in medical imaging, skin cancer detection, and associated smartphone apps and a web-based user-friendly diagnosis interface. We tested many available open-source ML algorithm-based software sets in Python as applied to medical image data processing, and modeling used to predict cancer growths and treatments. We study the use of ML concepts that promote efficient, accurate, secure computation over medical images, identifying and classifying cancer cells, and modeling the cancer cell growths. In this collaborative project with another university, we follow a holistic approach to data analysis leading to more efficient cancer detection based upon both cell analysis and image recognition. Here, we compare ML based software methods and analyze their detection accuracy. In addition, we acquire publicly available data of cancer cell image files and analyze using deep learning algorithms to detect benign and suspicious image samples. We apply the current pattern matching algorithms and study the available data with possible diagnosis of cancer types.more » « less
-
Traditional fingerprint authentication requires the acquisition of data through touch-based specialized sensors. However, due to many hygienic concerns including the global spread of the COVID virus through contact with a surface has led to an increased interest in contactless fingerprint image acquisition methods. Matching fingerprints acquired using contactless imaging against contact-based images brings up the problem of performing cross modal fingerprint matching for identity verification. In this paper, we propose a cost-effective, highly accurate and secure end-to-end contactless fingerprint recognition solution. The proposed framework first segments the finger region from an image scan of the hand using a mobile phone camera. For this purpose, we developed a cross-platform mobile application for fingerprint enrollment, verification, and authentication keeping security, robustness, and accessibility in mind. The segmented finger images go through fingerprint enhancement to highlight discriminative ridge-based features. A novel deep convolutional network is proposed to learn a representation from the enhanced images based on the optimization of various losses. The proposed algorithms for each stage are evaluated on multiple publicly available contactless databases. Our matching accuracy and the associated security employed in the system establishes the strength of the proposed solution framework.more » « less
-
Abstract The boundary of solar system object discovery lies in detecting its faintest members. However, their discovery in detection catalogs from imaging surveys is fundamentally limited by the practice of thresholding detections at signal-to-noise (SNR) ≥ 5 to maintain catalog purity. Faint moving objects can be recovered from survey images using the shift-and-stack algorithm, which coadds pixels from multi-epoch images along a candidate trajectory. Trajectories matching real objects accumulate signal coherently, enabling high-confidence detections of very faint moving objects. Applying shift-and-stack comes with high computational cost, which scales with target object velocity, typically limiting its use to searches for slow-moving objects in the outer solar system. This work introduces a modified shift-and-stack algorithm that trades sensitivity for speedup. Our algorithm stacks low-SNR detection catalogs instead of pixels, the sparsity of which enables approximations that reduce the number of stacks required. Our algorithm achieves real-world speedups of 10–103× over image-based shift-and-stack while retaining the ability to find faint objects. We validate its performance by recovering synthetic inner and outer solar system objects injected into images from the DECam Ecliptic Exploration Project. Exploring the sensitivity–compute time trade-off of this algorithm, we find that our method achieves a speedup of ∼30× with 88% of the memory usage while sacrificing 0.25 mag in depth compared to image-based shift-and-stack. These speedups enable the broad application of shift-and-stack to large-scale imaging surveys and searches for faint inner solar system objects. We provide a reference implementation via thefind-asteroidsPython package and this URL:https://github.com/stevenstetzler/find-asteroids.more » « less
-
With rich visual data, such as images, becoming readily associated with items, visually-aware recommendation systems (VARS) have been widely used in different applications. Recent studies have shown that VARS are vulnerable to item-image adversarial attacks, which add human-imperceptible perturbations to the clean images associated with those items. Attacks on VARS pose new security challenges to a wide range of applications, such as e-commerce and social media, where VARS are widely used. How to secure VARS from such adversarial attacks becomes a critical problem. Currently, there is still a lack of systematic studies on how to design defense strategies against visual attacks on VARS. In this article, we attempt to fill this gap by proposing anadversarial image denoising and detectionframework to secure VARS. Our proposed method can simultaneously (1) secure VARS from adversarial attacks characterized bylocalperturbations by image denoising based onglobalvision transformers; and (2) accurately detect adversarial examples using a novel contrastive learning approach. Meanwhile, our framework is designed to be used as both a filter and a detector so that they can bejointlytrained to improve the flexibility of our defense strategy to a variety of attacks and VARS models. Our approach is uniquely tailored for VARS, addressing the distinct challenges in scenarios where adversarial attacks can differ across industries, for instance, causing misclassification in e-commerce or misrepresentation in real estate. We have conducted extensive experimental studies with two popular attack methods (FGSM and PGD). Our experimental results on two real-world datasets show that our defense strategy against visual attacks is effective and outperforms existing methods on different attacks. Moreover, our method demonstrates high accuracy in detecting adversarial examples, complementing its robustness across various types of adversarial attacks.more » « less
