Teaching-Learning-based Optimization is an optimization technique which does not require any algorithm-specific parameters and is popular for its less computational cost and high consistency. Therefore, it has achieved great success application by the researchers in various disciplines of engineering. It works on the philosophy of teaching and learning which is used to solve multi-dimensional, linear and nonlinear problems with appreciable efficiency. Recently the basic TLBO algorithm is improved to enhance its exploration and exploitation capacities and the performance. However, there is less surveys on TLBO algorithm recent advances and its application. In this paper, the successful researches of TLBO algorithm of the past decade are surveyed. Firstly, the available intelligent optimization algorithms were reviewed. Then the application fields of TLBO and the improved TLBO were discussed and analyzed. Furthermore, some representative TLBO methods were classified into three main groups: 1) Improvement of teaching process; and 2) Fusion with Other Optimization Methods; and 3) Weight Methods and Others. Finally, our viewpoints were shared on the open issues and challenges in TLBO as well as research trends in the future.
Aiming at the problem of the weak avalanche effect in the recently proposed deep learning image encryption algorithm, this paper analyzes the causes of weak avalanche effect in the neural network of Cycle-GAN step-by-step process and proposes an image encryption algorithm combining the traditional diffusion algorithm and deep learning neural network. In this paper, first, the neural network is used for image scrambling and slight diffusion, and then the traditional diffusion algorithm is used to further diffuse the pixels. The experiment in satellite images shows that our algorithm, with the help of the further diffusion mechanism, can compensate for the weak avalanche effect of Cycle-GAN-based image encryption and can change a pixel value to the original image, and the number of pixel change rate (NPCR) and unified average changing intensity (UACI) values can achieve 99.64% and 33.49%, respectively. In addition, our method can effectively encrypt the image where the encrypted image with high information entropy and low pixel correlation is obtained. The experiment on data loss and noise attack declares our method can identify the types and intensity of attacks. What is more, the key space is big enough, and the key sensitivity is high while the key has a certain randomness.
Vanishing-point detection is an important component for the visual navigation system of an autonomous mobile robot. In this paper, we present a novel line space voting method for fast vanishing-point detection. First, the line segments are detected from the road image by the line segment detector (LSD) method according to the pixel’s gradient and texture orientation computed by the Sobel operator. Then, the vanishing-point of the road is voted on by considering the points of the lines and their neighborhood spaces with weighting methods. Our algorithm is simple, fast, and easy to implement with high accuracy. It has been experimentally tested with over hundreds of structured and unstructured road images. The experimental results indicate that the proposed method is effective and can meet the real-time requirements of navigation for autonomous mobile robots and unmanned ground vehicles.
In such a highly information-oriented era, digital image transmission and delivery on the Internet have been increasingly frequent. the deliverer hopes that this only occurs in a secure channel while on the Internet; however; there are many eavesdroppers and destroyers. To protect individual privacy on public network platforms, researchers need to find an approach that satisfies both private image security and robustness. Image encryption, image steganography, and image authentication are three efficient methods that balance the characteristics of images and the requirement of security. Deep learning is a powerful tool in image processing that has reached impressive successes in image object detection,1–3 image classification,6–9 image segmentation,10–13 image style transfer,14–17 image denoising,18–20 and image compression.21–23 Applying deep learning methods to the field of image security to solve the traditional problems has also received extensive attention and achieved breakthrough progress recently. But how to better use the advantages of deep learning in image steganography, image cryptography, and image authentication always attracts many scholars’ attention. To help relevant researchers understand the field of deep learning applications in digital image security and its future development more quickly, in this paper, we order the origin and development process of deep learning methods in image steganography, cryptography, and authentication from multiple aspects, as can be seen in Fig. 1; we then compare these methods, analyze the advantages and disadvantages of each, and finally suggest future research directions of this field.