Micro-Expression Recognition Based on Pixel Residual Sum and Cropped Gaussian Pyramid

Facial micro-expression(ME) recognition has great significance for the progress of human society and could find a person's true feelings. Meanwhile, ME recognition faces a huge challenge, since it is difficult to detect and easy to be disturbed by the environment. In this article, we propose two novel preprocessing methods based on Pixel Residual Sum. These methods can preprocess video clips according to the unit pixel displacement of images, resist environmental interference, and be easy to extract subtle facial features. Furthermore, we propose a Cropped Gaussian Pyramid with Overlapping(CGPO) module, which divides images of different resolutions through Gaussian pyramids and crops different resolutions images into multiple overlapping subplots. Then, we use a convolutional neural networks of progressively increasing channels based on the depthwise convolution to extract preliminary features. Finally, we fuse preliminary features and make position embedding to get the last features. Our experiments show that the proposed methods and model have better performance than the well-known methods.

Design of Multi-Receptive Field Fusion-Based Network for Surface Defect Inspection on Hot-Rolled Steel Strip Using Lightweight Dataset

With the advancement of industrial intelligence, defect recognition has become an indispensable part of facilitating surface quality in the steel manufacturing process. To assure product quality, most previous studies were typically trained with many defect samples. Nonetheless, a large quantity of defect samples is difficult to obtain, owing to the rare occurrence of defects. In general, deep learning-based methods underperformed as they have inherent limitations due to inadequate information, thereby restraining the application of models. In this study, a two-level Gaussian pyramid is applied to decompose raw data into different resolution levels simultaneously filtering the noises to acquire compact and representative features. Subsequently, a multi-receptive field fusion-based network (MRFFN) is developed to learn the hierarchical features and synthesize the respective prediction scores to form the final recognition result. As a result, the proposed method is capable of exhibiting an outstanding performance of 99.75% when trained using a lightweight dataset. In addition, the experiments conducted using the disturbance defect dataset showed the robustness of the proposed MRFFN against common noises and motion blur.

A Rotation-Invariant Optical and SAR Image Registration Algorithm Based on Deep and Gaussian Features

Traditional feature matching methods of optical and synthetic aperture radar (SAR) used gradient are sensitive to non-linear radiation distortions (NRD) and the rotation between two images. To address this problem, this study presents a novel approach to solving the rigid body rotation problem by a two-step process. The first step proposes a deep learning neural network named RotNET to predict the rotation relationship between two images. The second step uses a local feature descriptor based on the Gaussian pyramid named Gaussian pyramid features of oriented gradients (GPOG) to match two images. The RotNET uses a neural network to analyze the gradient histogram of the two images to derive the rotation relationship between optical and SAR images. Subsequently, GPOG is depicted a keypoint by using the histogram of Gaussian pyramid to make one-cell block structure which is simpler and more stable than HOG structure-based descriptors. Finally, this paper designs experiments to prove that the gradient histogram of the optical and SAR images can reflect the rotation relationship and the RotNET can correctly predict them. The similarity map test and the image registration results obtained on experiments show that GPOG descriptor is robust to SAR speckle noise and NRD.

Real Time Night Vision Surveillance using Improved Dark Channel Prior

Videos taken under low lighting conditions usually result in severe loss of visibility and contrast and are uncomfortable for observation and analysis. Night vision cameras that cater to the needs are expensive and less versatile. To be cost effective and extract maximum information from videos taken in low lit conditions, video enhancing techniques must be used. Though there are many night vision enhancement techniques available in literature, this paper particularly emphasizes about Improved Dark Channel Prior algorithm and its results. This approach suits well for real time night video enhancement. It has been found that a pixel-wise inversion of a night video appears very similar to the video obtained during foggy days. The same idea of haze removal approach is used to boost the visual quality of night videos. An improved dark channel prior model is presented that is integrated with Gaussian Pyramid operators for local smoothing. The experimental results show that the proposed method can boost the perceptual quality of detailing in night videos.

Infrared and visible image fusion via octave Gaussian pyramid framework

Image fusion integrates information from multiple images (of the same scene) to generate a (more informative) composite image suitable for human and computer vision perception. The method based on multiscale decomposition is one of the commonly fusion methods. In this study, a new fusion framework based on the octave Gaussian pyramid principle is proposed. In comparison with conventional multiscale decomposition, the proposed octave Gaussian pyramid framework retrieves more information by decomposing an image into two scale spaces (octave and interval spaces). Different from traditional multiscale decomposition with one set of detail and base layers, the proposed method decomposes an image into multiple sets of detail and base layers, and it efficiently retains high- and low-frequency information from the original image. The qualitative and quantitative comparison with five existing methods (on publicly available image databases) demonstrate that the proposed method has better visual effects and scores the highest in objective evaluation.