Image Deblurring of Video Surveillance System in Rainy Environment

Video surveillance system is used in various fields such as transportation and social life. The bad weather can lead to the degradation of the video surveillance image quality. In rainy environment, the raindrops and the background are mixed, which lead to make the image degradation, so the removal of the raindrops has great significance for image restoration. In this article, after analyzing the inter-frame difference method in detecting and removing raindrops, a background difference method is proposed based on Gaussian model. In this method, the raindrop is regarded as a moving object relative to the background. The principle and procedure of the method are given to detect and remove raindrops. The parameters of the single Gaussian background model are studied in this article. The important parameter of the learning rate of Gaussian model is explored in order to better detection and removal of raindrops. Experiment shows that the results of removal of raindrops by using the proposed algorithm are better than that by using the inter-frame difference method. The image processing effect is the best when the learning rate is 0.6. The research results can provide technical reference for similar research on eliminating the influence of rainy weather.     

考虑任务拆分特性与簇准备时间的并行机调度

在印制电路板钻孔任务调度等工程实际中,普遍存在一类具有任务拆分特性与簇准备时间的并行机调度问题,尚缺乏高效的优化模型和方法。针对该问题,首先建立以总拖期最小为目标的数学模型,以约束的形式将两个现有优势定理嵌入其中。为了高效求解实际规模问题,进一步提出嵌入优势定理的模拟退火算法。最后,基于随机生成的算例构造计算实验,以验证所建模型和算法的有效性。实验结果表明,嵌入优势定理的数学模型在问题求解规模和计算效率方面均优于现有数学模型,嵌入优势定理的模拟退火算法同样优于现有模拟退火算法。     

对光照变化鲁棒的一种运动目标检测方法

鉴于现有的运动目标检测方法对光照变化的敏感性,本文提出了一种在静态场景下对光照变化鲁棒的运动目标检测方法。该方法利用视频当前图像帧的边缘与背景图像的边缘做差分运算,得出运动目标的轮廓,进而对运动目标进行定位与检测。实验证明,本方法在一定程度上能够消除由于环境光照变化引起的"曝光"现象,实时准确地检测出运动目标及其位置,运算速度快,满足实时性的需求。     

A Deep Learning-Based Crowd Counting Method and System Implementation on Neural Processing Unit Platform

In this paper, a deep learning-based method is proposed for crowd-counting problems. Specifically, by utilizing the convolution kernel density map, the ground truth is generated dynamically to enhance the feature-extracting ability of the generator model. Meanwhile, the “cross stage partial” module is integrated into congested scene recognition network (CSRNet) to obtain a lightweight network model. In addition, to compensate for the accuracy drop owing to the lightweight model, we take advantage of “structured knowledge transfer” to train the model in an end-to-end manner. It aims to accelerate the fitting speed and enhance the learning ability of the student model. The crowd-counting system solution for edge computing is also proposed and implemented on an embedded device equipped with a neural processing unit. Simulations demonstrate the performance improvement of the proposed solution in terms of model size, processing speed and accuracy. The performance on the Venice dataset shows that the mean absolute error (MAE) and the root mean squared error (RMSE) of our model drop by 32.63% and 39.18% compared with CSRNet. Meanwhile, the performance on the ShanghaiTech PartB dataset reveals that the MAE and the RMSE of our model are close to those of CSRNet. Therefore, we provide a novel embedded platform system scheme for public safety pre-warning applications.     

Real-time multi-face detection on FPGA for video surveillance applications

In the field of face recognition and in the establishment of a face database, face detection is a crucial step. In current security and surveillance systems, most of the face detection proposed now is focused on software algorithms to improve the detection rate and decrease false alarms. However, these more complex algorithms require more computation time, which hinders real-time applications. In this paper, we propose a real-time multi-face detection system based on hardware design to enhance processing time. The proposed hardware architecture is implemented on an Altera DE2-70 field-programmable gate array development board to test the feasibility of our hardware design. To implement, our system requires 15,223 logic elements. The proposed system can operate in real-time at a frame rate of 30?fps, and detects up to five faces simultaneously. Our experimental results show that the proposed face detection architecture provides a reliable real-time system operating at low cost and providing a high detection rate.     

无人机电子稳像技术中角点检测算法的改进

无人机航空遥感电子稳像系统中,稳像的关键技术之一是影像特征点的选取,其中图像角点是遥感影像中重要的特征信息,准确地选取角点可提高图像处理的精度。然而现有的图像角点检测算法多因计算速度慢不能满足视频图像数字稳像的实时性。因此提出了一种基TSUSAN角点检测算法的改进算法。新算法分析了影像中角点所在区域的灰度变化特征,改进了SUSAN角点检测算法中的判断准则,提高了算法的精度和速度。实验结果表明,改进的算法可较大幅度的提高运算速度,满足稳像技术对视频图像实时处理的要求。     

An Image Steganography Algorithm Based on Quantization Index Modulation Resisting Scaling Attacks and Statistical Detection

In view of the fact that the current adaptive steganography algorithms are difficult to resist scaling attacks and that a method resisting scaling attack is only for the nearest neighbor interpolation method, this paper proposes an image steganography algorithm based on quantization index modulation resisting both scaling attacks and statistical detection. For the spatial image, this paper uses the watermarking algorithm based on quantization index modulation to extract the embedded domain. Then construct the embedding distortion function of the new embedded domain based on S-UNIWARD steganography, and use the minimum distortion coding to realize the embedding of the secret messages. Finally, according to the embedding modification amplitude of secret messages in the new embedded domain, the quantization index modulation algorithm is applied to realize the final embedding of secret messages in the original embedded domain. The experimental results show that the algorithm proposed is robust to the three common interpolation attacks including the nearest neighbor interpolation, the bilinear interpolation and the bicubic interpolation. And the average correct extraction rate of embedded messages increases from 50% to over 93% after 0.5 times-fold scaling attack using the bicubic interpolation method, compared with the classical steganography algorithm S-UNIWARD. Also the algorithm proposed has higher detection resistance than the original watermarking algorithm based on quantization index modulation.     

A VIBRATION RECOGNITION METHOD BASED ON DEEP LEARNING AND SIGNAL PROCESSING

Effective vibration recognition can improve the performance of vibration control and structural damage detection and is in high demand for signal processing and advanced classification.Signal-processing methods can extract the potent time-frequency-domain characteristics of signals;however,the performance of conventional characteristics-based classification needs to be improved.Widely used deep learning algorithms(e.g.,convolutional neural networks(CNNs))can conduct classification by extracting high-dimensional data features,with outstanding performance.Hence,combining the advantages of signal processing and deep-learning algorithms can significantly enhance vibration recognition performance.A novel vibration recognition method based on signal processing and deep neural networks is proposed herein.First,environmental vibration signals are collected;then,signal processing is conducted to obtain the coefficient matrices of the time-frequency-domain characteristics using three typical algorithms:the wavelet transform,Hilbert-Huang transform,and Mel frequency cepstral coefficient extraction method.Subsequently,CNNs,long short-term memory(LSTM)networks,and combined deep CNN-LSTM networks are trained for vibration recognition,according to the time-frequencydomain characteristics.Finally,the performance of the trained deep neural networks is evaluated and validated.The results confirm the effectiveness of the proposed vibration recognition method combining signal preprocessing and deep learning.     

改进式背景差分算法研究

该文提出一种改进式背景差分算法,并应用于监控系统中。针对人流量较少的监控情况,提出一种基于计算机视觉的嵌入式监控系统解决方案,其以DSP DM642为核心处理芯片,可对3路视频视角同时处理。系统利用动态权值的改进式背景差分算法对视频流进行实时监控,若发现异常事件,则自动存储一段时间的视频数据,并利用H.264压缩后保存至外存中以供事后取证。由于监控算法巧妙,普通的SD卡即可替代传统的硬盘,系统精简,使用方便。试验表明:该系统灵敏度可调,非常适用于外景和内景的库房监控。     

基于COTS处理器的微小卫星软件容错策略研究

为解决基于COTS处理器的现代微小卫星系统的软件容错问题,提出了基于虚拟寄存器的软件加固技术(SHVR),它把运行在COTS处理器上的软件故障分为数据流错误和控制流错误,设计了一套虚拟寄存器分配方法来突破现有的只能采用高级语言源程序作为输入的限制,提出了基于虚拟寄存器的数据流和控制流错误检测算法,并在实际应用背景下对这一方案进行了优化.模拟实验和实际应用表明,这套完全基于软件技术的方案在平均付出82.6%性能代价的前提下,对随机注入故障检测率达到91.4%.该方法现已成功应用于哈工大某重大航天课题中.     

Nonstationary weak signal detection based on normalization stochastic resonance with varying parameters

The nonlinear stochastic resonance system possesses the ability of taking advantage of background noise to enhance the weak signal. It provides a new approach to detect the weak signal embedded with heavy noise. This study proposes a new varying parameter stochastic resonance employing the fourth-order Runge–Kutta numerical method as well as the normalized transformation of a bistable stochastic resonance system. The model performs well in the detection of a time-varying signal with background noise for denoising and signal recovery. We take the fitness coefficient and cross-correlation coefficient as the criteria and analyze the influence of different parameters. The simulating results indicate its availability, validity and that it generates a better performance than the traditional stochastic resonance. The method develops the area of time-varying signal detection with stochastic resonance and presents new strategy for detection and denoising of a time-varying signal. It can be expected to be widely used in the areas of aperiodic signal processing, radar communication, etc.     

基于图像序列的运动目标检测识别的关键技术研究

为提高运动目标检测的识别效果,通过分析、综合比较各种运动目标检测算法的优劣性,提出了基于全局自适应帧差法和基于码本模型的背景减除法对同一运动目标进行检测。通过对运动目标检测提取运动目标的掩膜,对掩膜进行外接矩形分析,从而得到包围运动目标的矩形框;将矩形框内的图片截取出来,调整该矩形并提取图片的HOG特征,最后通过训练好的SVM进行分类。在训练过程中,针对难易情况应用自举法对训练器进行优化。实验表明,与传统HOG+SVM多尺度检测算法相比,该方法在速度和准确性上可提升20%左右,可作为运动目标检测与识别的参考方法。     

The algorithm of accelerated cracks detection and extracting skeleton by direction chain code in concrete surface image

Due to concrete surface roughness, uneven illumination, shadows, complex background and other disruptive factors, the traditional image processing-based concrete crack detection method cannot accurately detect concrete cracks, especially unclear ones and some tiny ones. The crack detection method based on the percolation model, which fully considered the low brightness and slenderness of the cracks, can accurately detect unclear and tiny cracks. But this method is time-consuming, and in some cases, it may cause fractures on the detected cracks. In order to solve these problems, this paper proposed an improved algorithm of image crack inspection based on the percolation model, which can reduce processing time through reducing the number of percolated pixels. To reconnect the fractured cracks, this method extracts the skeleton of cracks first by using an algorithm of skeleton extraction based on direction chain code. Then this paper proposed a region extension-based algorithm to reconnect part of the fractured cracks. Experimental results showed that this algorithm can significantly accelerate crack detection and maintain high detection precision.     

基于深度学习的合作目标靶球检测

为了解决复杂场景下激光跟踪仪对合作目标靶球的精确识别难题,提出了基于深度学习的合作目标靶球高效检测方法。首先分析了合作目标靶球的图像特征,然后采用改进的YOLOv2模型,针对合作目标靶球多尺度与小目标占比多的特点,提出了一种基于注意力机制的改进方法,同时为提高网络模型对复杂背景的抗干扰能力,提出了一种数据增强方法。测试结果表明,所提出的基于注意力机制与数据增强的改进YOLOv2模型对复杂背景的抗干扰能力较强,且对合作目标靶球的检测精度有显著提高,在合作目标靶球测试集上的检测准确率达到92.25%,能够有效满足激光跟踪仪在大型装置精密装配过程中的目标检测精度需求。     

设计草图中相机定标技术的研究

以基于设计草图的3D模型检索技术为应用背景,提出了快速有效的灭点探测和相机定标方法.首先分析了现有灭点探测方法的优缺点,从整体最优化和快速探测出发,得出了适合基于设计草图进行检索的草图灭点探测方法;然后对相机定标技术进行研究,对线性定标模型进行改进,得出了快速有效的相机定标方法.     

RETRACTED: Implementation System of Human Eye Tracking Algorithm Based on FPGA

With the high-speed development of transportation industry, highway traffic safety has become a considerable problem. Meanwhile, with the development of embedded system and hardware chip, in recent years, human eye detection eye tracking and positioning technology have been more and more widely used in man-machine interaction, security access control and visual detection.
In this paper, the high parallelism of FPGA was utilized to realize an elliptical approximate real-time human eye tracking system, which was achieved by the series register structure and random sample consensus (RANSAC), thus improving the speed of image processing without using external memory. Because eye images acquired by the camera often generate a lot of noises due to uneven light and dark background, the preprocessing technologies such as color conversion, image filtering, histogram modification and image sharpening were adopted. In terms of feature extraction of images, the eye tracking algorithm in this paper adopted seven-section rectangular eye tracking characteristic method, which increased a section between the mouth and the nose on the basis of the traditional six-section method, so its recognition accuracy is much higher. It is convenient for the realization of hardware parallel system in FPGA. Finally, aiming at the accuracy and real-time performance of the design system, a more comprehensive simulation test was carried out.
The human eye tracking system was verified on DE2-115 multimedia development platform, and the performance of VGA (resolution: 640×480) images of 8-bit grayscale was tested. The results showed that the detection speed of this system was about 47 frames per second under the condition that the detection rate of human face (front face, no inclination) was 93%, which reached the real-time detection level. Additionally, the accuracy of eye tracking based on FPGA system was more than 95%, and it has achieved ideal results in real-time performance and robustness.     

多方法融合技术在多目标位置检测中的应用

提出了将多种技术融合一体定位多目标的方法。首先根据计算机视觉原理，采用正交光轴的双摄像机配置检测系统；然后建立去除复杂实际背景和自适应二值化的动态图象预处理模型，并根据矩法搜索和求取目标质心；再利用神经网络模拟人眼感知事物的功能，确立图象点与空间点的映射关系，并将目标质心象点坐标作为输入节点，采用两个BP网并行处理左右两幅图象，快速求得目标空间位置。实践表明，此方法具有一定的实用性，并且在实际应用中取得了满意的效果。     

基于YOLO模型的机器人电梯厅门装箱状态快速识别方法

目的针对电梯厅门柔性生产线机器人装箱后厅门状态识别问题,提出一种基于YOLO模型的电梯厅门装箱状态快速识别方法。方法采用工业相机采集装箱后厅门图像信息,并制作成样本训练集,然后将训练集输入到目标识别分类检测模型中,通过调整网络结构参数进行迭代训练。结果经过测试验证,文中提出的识别方法对装箱后厅门的状态分类识别成功率在99%以上,而且识别速度明显优于传统机器视觉处理算法。结论文中提出的厅门装箱状态快速识别方法,可有效解决工业环境中复杂多变光照因素对传统机器视觉处理算法造成的识别效率低、误判率高等问题,并能满足生产系统节拍要求。     

A cost- and energy-efficient embedded system design based on instruction execution frequencies

In this paper, an embedded system that takes into account the frequency of the executed instructions to reduce memory space and save energy consumption is proposed. The object codes are split into the frequently executed instructions and the infrequently executed instructions (INFIs) by analyzing the trace files of applications. To reduce the use of memory space, the dictionary-based method is used to compress INFIs. To take into account energy consumption, the top-executed instructions are selected to encode as shorter codewords and wrapped into a pseudo instruction. When a pseudo instruction that contains several codewords is fetched, it can be decompressed to several continuous instructions to reduce the number of memory accesses. In addition, to further reduce energy consumption, a multiple reference table design is proposed to make a pseudo instruction contain more encoded codewords by shortening the length of an encoded instruction. From the simulation results, the proposed design that uses one 256-instruction reference table reduces the energy consumption about 50.4% compared to the dictionary-based method. In addition, to show the improvement of energy consumption for the proposed multiple reference table method over that using one reference table, we also show the simulation results of a design with two 256-instruction reference tables which shows less energy use than a design with one 512-instruction reference table by 12.1%.     

Image reconstruction in photoacoustic tomography involving layered acoustic media

Photoacoustic tomography (PAT), also known as thermoacoustic or optoacoustic tomography, is a rapidly emerging biomedical imaging technique that combines optical image contrast with ultrasound detection principles. Most existing reconstruction algorithms for PAT assume the object of interest possesses homogeneous acoustic properties. The images produced by such algorithms can contain significant distortions and artifacts when the object's acoustic properties are spatially variant. In this work, we establish an image reconstruction formula for PAT applications in which a planar detection surface is employed and the to-be-imaged optical absorber is embedded in a known planar layered acoustic medium. The reconstruction formula is exact in a mathematical sense and accounts for multiple acoustic reflections between the layers of the medium. Computer-simulation studies are conducted to demonstrate and investigate the proposed method.