Image and Feature Space Based Domain Adaptation for Vehicle Detection

The application of deep learning in the field of object detection has experienced much progress. However, due to the domain shift problem, applying an off-the-shelf detector to another domain leads to a significant performance drop. A large number of ground truth labels are required when using another domain to train models, demanding a large amount of human and financial resources. In order to avoid excessive resource requirements and performance drop caused by domain shift, this paper proposes a new domain adaptive approach to cross-domain vehicle detection. Our approach improves the cross-domain vehicle detection model from image space and feature space. We employ objectives of the generative adversarial network and cycle consistency loss for image style transfer in image space. For feature space, we align feature distributions between the source domain and the target domain to improve the detection accuracy. Experiments are carried out using the method with two different datasets, proving that this technique effectively improves the accuracy of vehicle detection in the target domain.     

提高关联成像重构质量的研究

比较基于压缩感知关联成像(CGI)与伪逆关联成像(PGI)两者之间的成像效果差异,探讨形态学权重自适应对关联成像去除噪声的效果。选择不同的图像,通过MATLAB软件开展仿真实验,对目标图像分别采样64、256、512、1 024、2 048、3 000次,首先通过关联成像、基于压缩感知关联成像与伪逆关联成像三种方法重构图像,再对比压缩感知与伪逆两种方法重构图像的效果,以峰值信噪比(PSNR)、相关系数(CC)为量化指标,将基于压缩感知关联成像与伪逆关联成像在不同采样次数下进行对比分析。同时,通过实验分析形态学权重自适应去除关联成像中噪声的效果。伪逆关联成像在低次数采样的情况下比基于压缩感知关联成像的成像效果更好,在高采样次数下,基于压缩感知关联成像的成像效果更好。在实际重构中压缩感知关联成像重构的图像仍有噪声,形态学权重自适应可以有效去除关联成像实验中产生的噪声。     

基于机器视觉的手机电池表面缺陷检测

针对手机电池表面质量人工检测情况,开发了电池表面缺陷无损检测系统软件。首先电池表面经过倾斜矫正、感兴趣区域提取和字符灰度值修改等预处理操作,通过基于灰度密度分布和灰度差的自适应阈值亮度法对感兴趣区域进行子图像遍历,融合有重合区域的缺陷子图像并滤除没有明显缺陷的区域;然后采用支持向量机多种类分类法,提取二值图像像素分布规律作为训练特征,识别电池表面缺陷种类;最后设计了人机交互界面,确定最佳的可变参数,实验测试缺陷识别率达95%以上。     

Occurrence and diversity of Listeria spp. in seafood processing plant environments

Listeria contamination in processing plant environments is a major issue for the seafood industry worldwide; faster and more reliable results are therefore desired for early detection and monitoring of environmental Listeria spp. This study aimed to gain a better understanding of the prevalence and diversity of Listeria spp., and to evaluate a rapid detection method, the 3M Molecular Detection Assay (MDA) Listeria, for its ability to detect Listeria spp. in environmental samples from seafood processing plants. Duplicate environmental sponge samples (n = 444) were collected from 152 different sites within three seafood processing plants, and analyzed for Listeria spp. by the MDA method (after 26 and 48 h of enrichment) and the U.S. Food and Drug Administration Bacteriological Analytical Manual method. Overall, detection of Listeria spp. by the two methods did not differ significantly (p > 0.05); 11 (4.9%) and 13 (5.9%) samples were positive for Listeria spp. by the MDA and FDA-BAM method, respectively. The sensitivity of the MDS was 87.0% (95% CI: 77.4–96.6%), specificity was 97.6% (95% CI: 95.5–99.7%), accuracy was 95.3%, and the positive predictive value was 89.4% (95% CI: 80.5–98.2%). Classification of 19 Listeria isolates by partial SigB sequencing analysis identified three allelic types. Twelve of these isolates were ATs 58 and 60 which were classified as Listeria monocytogenes lineage I and serotypes 1/2b, 3b, 4b, 4d, 4e, by multiplex-PCR serotyping. Six Listeria isolates were classified as Listeria innocua (AT31). Our data show that the 3M Molecular Detection Assay Listeria provides rapid and reliable results for detection and monitoring of Listeria spp., which are important for seafood processing plants. Effective Listeria monitoring programs will allow for improved development of Listeria control measures in order to minimize cross-contamination in finished products.     

Graph-based video fingerprinting using double optimal projection

A double optimal projection method that involves projections for intra-cluster and inter-cluster dimensionality reduction are proposed for video fingerprinting. The video is initially set as a graph with frames as its vertices in a high-dimensional space. A similarity measure that can compute the weights of the edges is then proposed. Subsequently, the video frames are partitioned into different clusters based on the graph model. Double optimal projection is used to explore the optimal mapping points in a low-dimensional space to reduce the video dimensions. The statistics and geometrical fingerprints are generated to determine whether a query video is copied from one of the videos in the database. During matching, the video can be roughly matched by utilizing the statistics fingerprint. Further matching is thereafter performed in the corresponding group using geometrical fingerprints. Experimental results show the good performance of the proposed video fingerprinting method in robustness and discrimination.     

Random-valued impulse noise removal using adaptive dual threshold median filter

Noise detection and its removal is very important in the image processing. Detection of noise is very crucial and significant in random valued impulse noise because it does not hamper the image pixels uniformly. This paper presents a novel and unique concept of adaptive dual threshold for the detection of random valued impulse noise along with simple median filter at noise removal stage. Simulation results shows that an efficient noise detection leads to a superior quality of de-noised image as compared to existing adaptive threshold based image de-noising techniques. Proposed threshold computation is based on averaging of pixel values of window which enhances the PSNR of our system as compared to existing median filter based image de-noising methods.     

Robust skin detection in real-world images

Human skin detection in images is desirable in many practical applications, e.g., human–computer interaction and adult-content filtering. However, existing methods are mainly suffer from confusing backgrounds in real-world images. In this paper, we try to address this issue by exploring and combining several human skin properties, i.e. color property, texture property and region property. First, images are divided into superpixels, and robust skin seeds and background seeds are acquired through color property and texture property of skin. Then we combining color, region and texture properties of skin by proposing a novel skin color and texture based graph cuts (SCTGC) to acquire the final skin detection results. Comprehensive and comparative experiments show that the proposed method achieves promising performance and outperforms many state-of-the-art methods over publicly available challenging datasets with a great part of hard images.     

碱化插层二维过渡金属碳化物的制备及其对铀酰离子的电化学检测

二维过渡金属碳化物(MXenes)具有良好的电化学性能与辐照稳定性, 其在放射性核素电化学检测领域有潜在应用价值。本研究通过碱活化的方式处理碳化钛型MXene(Ti₃C₂T_x), 随后将钾插层的Ti₃C₂T_x(K-Ti₃C₂T_x)负载到玻碳电极(GCE)上得到K-Ti₃C₂T_x/GCE修饰电极。采用XRD、SEM、XPS等手段分别对Ti₃C₂T_x和K-Ti₃C₂T_x进行分析表征, 并进一步研究了K-Ti₃C₂T_x/GCE对痕量铀酰离子(UO₂²⁺)的电化学检测性能。循环伏安(CV)实验结果表明, 相比于GCE电极, K-Ti₃C₂T_x/GCE修饰电极对UO₂²⁺的电化学响应显著增强。进一步使用差分脉冲伏安法(DPV)扫描, 发现pH=4.0时, K-Ti₃C₂T_x/GCE修饰电极对UO₂²⁺在铀浓度0.5~10 mg/L范围内呈现良好的线性检测关系, 本方法的检测限为0.083 mg/L(S/N=3), 稳定性和重复性好。     

A novel robot co-worker system for paint factories without the need of existing robotic infrastructure

This paper presents a human–robot co-working system to be applied to industrial tasks such as the production line of a paint factory. The aim is to optimize the picking task with respect to manual operation in a paint factory. The use of an agile autonomous robot co-worker reduces the time in the picking process of materials, and the reduction of the exposure time to raw materials of the worker improves the human safety. Moreover, the process supervision is also improved thanks to a better traceability of the whole process. The whole system consists of a manufacturing process management system, an autonomous navigation system, and a people detection and tracking system. The localization module does not require the installation of reflectors or visual markers for robot operation, significantly simplifying the system deployment in a factory. The robot is able to respond to changing environmental conditions such as people, moving forklifts or unmapped static obstacles like pallets or boxes. The system is not tied to specific manufacturing orders. It is fully integrated with the manufacturing process management system and it can process all possible orders as long as their components are placed into the warehouse. Real experiments to validate the system have been performed in a paint factory by a real holonomic platform and a worker. The results are promising from the evaluation of performance indicators such as exposure time of the worker to raw materials, automation of the process, robust and safe navigation, and the assessment of the end-user.