Accurate identification of road defects like potholes and cracks is essential for timely driver warnings, maintenance prioritization, and accident prevention. However, there is a gap in developing practical real-time road damage detection algorithms. This survey reviews the most efficient deep learning models and available Road Damage Detection (RDD) datasets, comparing them based on accuracy, complexity, and inference rate for real-time application suitability. This survey investigates the available datasets for road damage detection and the evaluation metrics used to assess object detection models. Additionally, it explores recent methods and deep learning models for small object detection, including You Only Look Once (YOLO), Region-based Convolutional Neural Network (R-CNN), and Single Shot Detector (SSD). The survey also provides an overview of Convolutional Neural Networks (CNN) fundamentals and attention mechanisms. A comparative analysis of the models and datasets is conducted to highlight their strengths and limitations. The key findings in road damage detection, particularly those related to the Crowdsensing-based Road Damage Detection Challenge (CRDDC, 2022), are summarized.
Multi-Secret Image Sharing (MSIS) is important in information security when multiple images are shared in an unintelligible form to different participants, where the images can only be recovered using the shares from participants. This paper proposes a simple and efficient (n, n)-MSIS system for colored images based on XOR and Chinese Remainder Theorem (CRT), where all the n share are required in the recovery. The system improves the security by adding dependency on the input images to be robust against differential attacks, and by using several delay units. It works with even and odd number of inputs, and has a long sensitive system key design for the CRT. Security analysis and a comparison with related literature are introduced with good results including statistical tests, differential attack measures, and key sensitivity tests as well as performance analysis tests such as time and space complexity. In addition, Field Programmable Gate Array (FPGA) realization of the proposed system is presented with throughput 530Mbits/sec. Finally, the proposed MSIS system is validated through software and hardware with all statistical analyses and proper hardware resources with low power consumption, high throughput and high level of security.
Multi-Secret Image Sharing (MSIS) is used when multiple images need to be shared to multiple participants, but the images can not be recovered without the presence of all shares. In this paper, a unified system for performing encryption and (n,n)-MSIS is proposed. While MSIS is based on the XOR operation, encryption combines the utilization of Chinese Remainder Theorem (CRT), SHA-256, and S-box for improved security. The same designed system is used for the generation of secret shares and the recovery of secret images. In addition, a sensitive system key is designed where three pairwise relatively prime subkeys are automatically generated for utilization in the CRT. The resulting secret shares pass statistical evaluation criteria such as RMSE, correlation, and entropy, and give good results for differential attack measures, and runtime. In addition, the proposed system succeeds in passing the NIST SP-800-22 statistical test suite and key sensitivity measures.
(n, n)-Visual Secret Sharing (VSS) allows a user to send an image in the form of shares to different participants. Every share can not reveal the secret alone, and only all shares together can reveal the secret with fast recovery. This paper proposes a pixel-based (n, n)-VSS system, where to share a pixel from the secret image, (n − 1) random pixels are generated from the Lorenz chaotic system for a varying set of (n − 1) shares. Then, the nth pixel is calculated for a random share using the secret pixel and the generated (n − 1) random pixels. The system is efficient, lossless, implemented for grayscale and color images, and has a simple XOR-based recovery scheme. It passed several security analysis tests and is robust against noise attacks. Moreover, performance analysis and comparisons with other VSS systems are presented, showing good results.
6 May 2024 in Journal of Real-Time Image Processing
Secret image sharing (SIS) conveys a secret image to mutually suspicious receivers by sending meaningless shares to the participants, and all shares must be present to recover the secret. This paper proposes and compares three systems for secret sharing, where a visual cryptography system is designed with a fast recovery scheme as the backbone for all systems. Then, an SIS system is introduced for sharing any type of image, where it improves security using the Lorenz chaotic system as the source of randomness and the generalized Arnold transform as a permutation module. The second SIS system further enhances security and robustness by utilizing SHA-256 and RSA cryptosystem. The presented architectures are implemented on a field programmable gate array (FPGA) to enhance computational efficiency and facilitate real-time processing. Detailed experimental results and comparisons between the software and hardware realizations are presented. Security analysis and comparisons with related literature are also introduced with good results, including statistical tests, differential attack measures, robustness tests against noise and crop attacks, key sensitivity tests, and performance analysis.