Underwater image super-resolution using multi-stage information distillation networks

Recently, single image super-resolution (SISR) has been widely applied in the fields of underwater robot vision and obtained remarkable performance. However, most current methods generally suffered from the problem of a heavy burden on computational resources with large model sizes, which limited their real-world underwater robotic applications. In this paper, we introduce and tackle the super resolution (SR) problem for underwater robot vision and provide an efficient solution for near real-time applications. We present a novel lightweight multi-stage information distillation network, named MSIDN, for better balancing performance against applicability, which aggregates the local distilled features from different stages for more powerful feature representation. Moreover, a novel recursive residual feature distillation (RRFD) module is constructed to progressively extract useful features with a modest number of parameters in each stage. We also propose a channel interaction & distillation (CI&D) module that employs channel split operation on the preceding features to produce two-part features and utilizes the inter channel-wise interaction information between them to generate the distilled features, which can effectively extract the useful information of current stage without extra parameters. Besides, we present USR-2K dataset, a collection of over 1.6K samples for large-scale underwater image SR training, and a testset with an additional 400 samples for benchmark evaluation. Extensive experiments on several standard benchmark datasets show that the proposed MSIDN can provide state-of-the-art or even better performance in both quantitative and qualitative measurements.     

WavNet — Visual saliency detection using Discrete Wavelet Convolutional Neural Network

In the recent advancements in image and video analysis, the detection of salient regions in the image becomes the initial step. This plays a crucial role in deciding the performance of such algorithms. In this work, a Multi-Resolution Feature Extraction (MRFE) technique that makes use of Discrete Wavelet Convolutional Neural Network (DWCNN) for generating features is employed. An Enhanced Feature Extraction (EFE) module extracts additional features from the high level features of the DWCNN, which are used to frame both channel as well as spatial attention models for yielding contextual attention maps. A new hybrid loss function is also proposed, which is a combination of Balanced Cross Entropy (BCE) loss and Edge based Structural Similarity (ESSIM) loss that effectively identifies and segments the salient regions with clear boundaries. The method is tested exhaustively with five different benchmark datasets and is proved superior to the existing state-of-the-art methods with a minimum Mean Absolute error (MAE) of 0.03 and F-measure of 0.956.     

基于光刀图像的机器人示教纠偏方法

为了解决机器人辅助在线检测系统示教过程中效率低、人为干预多的问题,针对孔类特征,提出一种通过处理光刀图像获取机器人位姿纠正参数的方法。在该方法中,纠偏过程按照将图像光刀线调整为水平方向、将被测特征调整至图像中央、根据不同特征调整最优扫描方向3个步骤进行。根据光刀图像可以求出测量坐标系与被测特征的位姿关系,并得到该位姿与理想测量位姿的偏差。经过坐标变换即可得到在机器人基坐标系下的调整参数。另外,可利用电脑控制机器人运动,从而实现了纠偏过程的自动化。实验结果表明,该方法可将初始的示教位姿修正为理想位姿,且比传统方式效率更高。     

Sub-band coding of hexagonal images

According to the circle-packing theorem, the packing efficiency of a hexagonal lattice is higher than an equivalent square tessellation. Consequently, in several contexts, hexagonally sampled images compared to their Cartesian counterparts are better at preserving information content. In this paper, novel mapping techniques alongside the wavelet compression scheme are presented for hexagonal images. Specifically, we introduce two tree-based coding schemes, referred to as SBHex (spirally-mapped branch-coding for hexagonal images) and BBHex (breadth-first block-coding for hexagonal images). Both of these coding schemes respect the geometry of the hexagonal lattice and yield better compression results. Our empirical results show that the proposed algorithms for hexagonal images produce better reconstruction quality at low bits per pixel representations compared to the tree-based coding counterparts for the Cartesian grid.     

基于改进残差特征蒸馏的轻量级超分辨率网络

基于深度学习的图像超分辨率算法通常采用递归的方式或参数共享的策略来减少网络参数，这将增加网络的深度，使得运行网络花费大量的时间，从而很难将模型部署到现实生活中。为了解决上述问题，本文设计一种轻量级超分辨率网络，对中间特征的关联性及重要性进行学习，且在重建部分结合高分辨率图像的特征信息。首先，引入层间注意力模块，通过考虑层与层之间的相关性，自适应地分配重要层次特征的权重。其次，使用增强重建模块提取高分辨率图像中更精细的特征信息，以此得到更加清晰的重建图片。通过大量的对比实验表明，本文设计的网络与其他轻量级模型相比，有更小的网络参数量，并且在重建精度和视觉效果上都有一定的提升。     

Improving accuracy of automatic optical inspection with machine learning

Electronic devices require the printed circuit board(PCB)to support the whole structure,but the assembly of PCBs suffers from welding problem of the electronic components such as surface mounted devices(SMDs)resistors.The automated optical inspection(AOI)machine,widely used in industrial production,can take the image of PCBs and examine the welding issue.However,the AOI machine could commit false negative errors and dedicated technicians have to be employed to pick out those misjudged PCBs.This paper proposes a machine learning based method to improve the accuracy of AOI.In particular,we propose an adjacent pixel RGB value based method to pre-process the image from the AOI machine and build a customized deep learning model to classify the image.We present a practical scheme including two machine learning procedures to mitigate AOI errors.We conduct experiments with the real dataset from a production line for three months,the experimental results show that our method can reduce the rate of misjudgment from 0.3%–0.5%to 0.02%–0.03%,which is meaningful for thousands of PCBs each containing thousands of electronic components in practice.     

Accelerating BPC-PaCo through Visually Lossless Techniques

Fast image codecs are a current need in applications that deal with large amounts of images. Graphics Processing Units (GPUs) are suitable processors to speed up most kinds of algorithms, especially when they allow fine-grain parallelism. Bitplane Coding with Parallel Coefficient processing (BPC-PaCo) is a recently proposed algorithm for the core stage of wavelet-based image codecs tailored for the highly parallel architectures of GPUs. This algorithm provides complexity scalability to allow faster execution at the expense of coding efficiency. Its main drawback is that the speedup and loss in image quality is controlled only roughly, resulting in visible distortion at low and medium rates. This paper addresses this issue by integrating techniques of visually lossless coding into BPC-PaCo. The resulting method minimizes the visual distortion introduced in the compressed file, obtaining higher-quality images to a human observer. Experimental results also indicate 12% speedups with respect to BPC-PaCo.     

高光谱成像用高帧频CMOS图像传感器北大核心

针对高帧频、全局曝光和光谱平坦等成像应用需求,设计了一款高光谱成像用CMOS图像传感器。其光敏元采用PN型光电二极管,读出电路采用5T像素结构。采用列读出电路以及高速多通道模拟信号并行读出的设计方案来获得低像素固定图像噪声(FPN)和非均匀性抑制。芯片采用ASMC 0.35μm三层金属两层多晶硅标准CMOS工艺流片,为了抑制光电二极管的光谱干涉效应,后续进行了光谱平坦化VAE特殊工艺,并对器件的光电性能进行了测试评估。电路测试结果符合理论设计预期,成像效果良好,像素具备积分可调和全局快门功能,最终实现的像素规模为512×256,像元尺寸为30μm×30μm,最大满阱电子为400 ke^(-),FPN小于0.2%,动态范围为72 dB,帧频为450 f/s,相邻10 nm波段范围内量子效率相差小于10%,可满足高光谱成像系统对CMOS成像器件的要求。     

基于光谱尺度空间与管道滤波的红外目标检测

为了准确地从复杂干扰背景下检测出真实弱小目标，本文引入视觉显著性，设计了基于快速光谱尺度空间与动态管道滤波的红外目标检测算法。基于真实目标与背景内容之间的整体差异，引入快速光谱尺度空间与阈值分割技术，设计视觉显著性机制，对红外图像完成处理，输出全局显著性映射，以高效过滤干扰背景内容。考虑目标与背景的局部特征差异，构建自适应局部对比度增强机制，对粗检测结果实施处理，获取对应的局部显著性映射，改善视觉显著性区域内目标的对比度。引入高斯差分理论，通过估算每一帧红外图像中的目标像素直径，形成动态管道滤波，充分消除虚警，准确识别出弱小目标。多组实验数据显示：较已有的红外目标检测技术而言，在各种不同的复杂背景干扰下，所提算法呈现出更好的检测能力，拥有更理想的接收机工作特性ROC曲线。