Application of fuzzy image restoration in criminal investigation

The advancement of science and technology has a positive effect on the development of law disciplines. The development of algorithms and artificial intelligence also has a certain impact on judicial practice. Image restoration is a significant technique in image processing. It aims to objectively restore the content or quality of the original image from the degraded image. Image degradation is always generated in image transmission, such as distortion, blur. In modern video surveillance system, image restoration is significant for criminal investigation. However, image restoration based on conventional filter algorithms cannot achieve satisfactory performance. Thus, we first introduce the image restoration algorithms based on different degradation model. Then, we propose some applications of fuzzy image restoration in criminal investigation. We conduct experiments on both degraded images and videos and experimental results have shown the effectiveness of fuzzy image restoration applying to the criminal investigation.     

Experimental investigation of multiphase flow behavior in drilling annuli using high speed visualization technique

Imaging with high definition video camera is an important technique to visualize the drilling conditions and to study the physics of complex multiphase flow associated with the hole cleaning process. The main advantage of visualizing multiphase flow in a drilling annulus is that the viewer can easily distinguish fluid phases, flow patterns and thicknesses of cutting beds. In this paper the hole cleaning process which involves the transportation of cuttings through a horizontal annulus was studied. The two-phase (solid-liquid) and the three-phase (solid-liquid-gas) flow conditions involved in this kind of annular transportation were experimentally simulated and images were taken using a high definition camera. Analyzing the captured images, a number of important parameters like velocities of different phases, heights of solid beds and sizes of gas bubbles were determined. Two different techniques based on an image analysis software and MATLAB coding were used for the determinations. The results were compared to validate the image analyzing methodology. The visualization technique developed in this paper has a direct application in investigating the critical conditions required for efficient hole cleaning as well as in optimizing the mud program during both planning and operational phases of drilling. Particularly, it would be useful in predicting the cuttings transport performance, estimating solid bed height, gas bubble size, and mean velocities of bubbles/particles.     

加热面朝下的池沸腾汽泡动态行为研究

基于Matlab软件开发了自动识别气液两相流界面程序，程序可获得气液界面变化、汽膜厚度、汽膜脱离周期和汽膜法向速度等特征。利用该程序对沟槽结构加热表面朝下布置时，在不同倾角、不同热流密度下的汽泡动态数据进行了处理和分析。结果表明：加热表面朝下发生核态沸腾时，汽膜厚度随热流密度的增大而增大，汽泡脱离周期随热流密度的增大先减小，而后维持在一稳定值；汽膜脱离周期随倾角的增大而减小，倾角为5°时的汽膜脱离周期稳定在0.27 s左右。当发生沸腾危机时，汽膜厚度迅速减小，这可作为动态监测加热表面沸腾状态的依据。     

一种用于平面的电火花强化修复装置

研究了一种用于金属平面的电火花强化修复装置。经过SEM照片分析,MM200磨损试验,以及硬度梯度分布综合判断,该装置生成的强化修复层比较完整、有一定厚度,极大改善了零件表面的硬度和耐磨性。     

自适应天牛须优化与K均值聚类的图像分割算法研究

在机器识别中,图像分割是重要的一个步骤,传统分割手段存在一定缺陷。针对传统K均值聚类分割的初始聚类中心敏感的缺陷进行了优化,利用自适应天牛须优化算法,避免了这一问题。通过实验结果表明,该算法(ABASK)对图像进行分割,既可以保证图像轮廓的分割,同时还可以更多地保留图像细节。     

A novel region-based multimodal image fusion technique using improved dictionary learning

Recently, the sparse representation (SR) based algorithms have gained much attention from the researchers in the area of image fusion (IF). The building of a compact discriminative dictionary plays a vital role in the sparse-based IF techniques. In this context, an efficient multimodal IF method based on improved dictionary learning is investigated. The key contributions of this paper are: (a) An improved KSVD algorithm is suggested for the dictionary learning process, (b) to reduce the computational time, only the informative patches are selected using energy feature, and (c) a novel region-based fusion scheme is suggested for the first time for the problem on hand. The suggested technique is tested with a number of multimodal images from Harvard Medical School brain database. The results are compared with state-of-the-art multiscale transform-based methods and modified SR-based methods. Unlike earlier methods, our proposed technique generates an adaptive dictionary through selection of informative patches only. This results in a compact dictionary with improved computational efficiency. The experimental results reveal that our approach outperforms other methods. The potential application of the suggested method could be in pathological images for follow-up study and better treatment planning.     

Dynamic Oscillation via Negative Differential Resistance in Type III Junction Organic/Two-Dimensional and Oxide/Two-Dimensional Transition Metal Dichalcogenide Diodes

Among many of 2D semiconductor-based devices, type III PN junction diodes are given special attentions due to their unique function, negative differential resistance (NDR). However, it has been found uneasy to achieve well-matched type III PN junctions from 2D–2D van der Waals heterojunctions. Here, the authors present other alternatives of type III heterojunctions, using 2D p-MoTe₂/organic n-type dipyrazino[2,3-f:2′,3′-h]quinoxaline-2,3,6,7,10,11-hexacarbonitrile (HAT-CN) and 2D p-WSe₂/n-MoO_x systems. Those junction diodes appear to well-demonstrate static and dynamic NDR behavior via resonant tunneling and electron–hole recombination. Extended to an inverter circuit, p-MoTe₂/n-HAT-CN diode enables multilevel inverter characteristics as monolithically integrated with p-MoTe₂ channel field effect transistor. The same NDR diode shows dynamic LC oscillation behavior under a constant DC voltage, connected to an external inductor. From p-WSe₂/n-MoO_x oxide diode, similar NDR behavior to those of p-MoTe₂/n-HAT-CN is again observed along with LC oscillations. The authors attribute these visible oscillation results to high peak-to-valley current ratios of their organic or oxide/2D heterojunction diodes.     

Discriminative graph convolution networks for hyperspectral image classification

Recently, the proposal of graph convolutional networks (GCN) has successfully implemented into hyperspectral image data representation and analysis. In spite of the great success, there are still several major challenges in hyperspectral image classification, including within-class diversity, and between-class similarity, which generally degenerate hyperspectral image classification performance. To address the problems, we propose a discriminative graph convolution networks (DGCN) for hyperspectral image classification. This method introduces the concepts of within-class scatter and between-class scatter, which respectively reflect the global geometric structure and discriminative information of the input space. The experimental results on the hyperspectral data sets show that the proposed method has good classification performance.     

The constrained two-dimensional guillotine cutting problem with defects: an ILP formulation,a Benders decomposition and a CP-based algorithm

This paper addresses a variant of two-dimensional cutting problems in which rectangular small pieces are obtained by cutting a rectangular object through guillotine cuts. The characteristics of this variant are (i) the object contains some defects, and the items cut must be defective-free; (ii) there is an upper bound on the number of times an item type may appear in the cutting pattern; (iii) the number of guillotine stages is not restricted. This problem commonly arises in industrial settings that deal with defective materials, e.g. either by intrinsic characteristics of the object as in the cutting of wooden boards with knotholes in the wood industry, or by the manufacturing process as in the production of flat glass in the glass industry. We propose a compact integer linear programming (ILP) model for this problem based on the discretisation of the defective object. As solution methods for the problem, we develop a Benders decomposition algorithm and a constraint-programming (CP) based algorithm. We evaluate these approaches through computational experiments, using benchmark instances from the literature. The results show that the methods are effective on different types of instances and can find optimal solutions even for instances with dimensions close to real-size.