License plate detection using gradient information and cascade detectors

In this paper, we propose a license plate detection method based on the gradient information and a cascade detection framework. The proposed method consists of three main modules: image preprocessing, license plate detection and license plate confirmation. Considering the diversities of Chinese license plate, we first get the gradient images to unify plate forms through image preprocessing. The license plate will be detected roughly by using the cascade AdaBoost classifier in the next step. Finally, several heuristic judgment strategies and a voting-based method are used to verify the candidate license plates. The effectiveness of our method is validated with an experimental study in license plate detection.     

基于自适应尺度的TLD目标跟踪算法

为提升TLD目标跟踪算法的每帧处理速度,以达到在更高分辨率视频中跟踪目标的实时性要求,在TLD算法框架的基础上,提出了一种基于自适应尺度检测学习的目标跟踪算法(AS-TLD)。当跟踪目标成功时,选取当前帧跟踪到的目标尺度及几个相邻的尺度作为下帧检测目标时滑动窗口尺度的选取范围;而当跟踪失败时,则选取在TLD算法初始化阶段,根据跟踪目标及视频图像大小选定的尺度来保障长时间跟踪目标,从而有效减少了平均每帧扫描的窗口数量。实验结果表明,该方法不仅有效地降低了检测模块的检测时间,显著提高了整体算法速度,而且通过动态选取尺度,在一定程度使得TLD各个模块更加协调,跟踪精确度得到提升。     

一种复杂车辆图像中的多车牌定位方法

针对复杂背景中多个车牌的定位问题,提出一种新的定位方法.该方法综合利用边缘检测、连通域分析、倾斜矫正等多种方法,解决了复杂背景中定位难的问题.能够准确定位杂乱背景中的车牌,对天气、光照变化、车牌在图像中的移动和旋转等,具有良好的适应能力.该方法为后续的字符分割和字符识别提供旋转角度、字符区域定位信息.     

基于TLD算法的加油锥套跟踪

针对自主空中加油对接阶段锥套跟踪问题,提出了一种基于tracking learning detection (TLD)的锥套跟踪算法。该算法将加油锥套的跟踪任务分解成跟踪、学习、检测3个部分。跟踪模块在LK光流法的基础上添加跟踪失败自检测,筛选出好的跟踪点,跟踪加油锥套;检测模块构建级联分类器,对滑动窗遍历得到的图像块进行分类并返回含有目标的图像块,融合跟踪模块的跟踪框,给出最终跟踪结果;学习模块引入P N约束修正错误样本并学习更新检测模块。利用Creator/Vega Prime软件对空中加油进行视景仿真,在视景仿真视频上测试锥套跟踪算法。结果表明:TLD算法跟踪加油锥套成功率达95.5%,处理每帧平均耗时31.4 ms,能够满足加油锥套跟踪鲁棒性、准确率、实时性的要求。     

基于BP神经网络的车牌字符识别算法研究

车牌字符识别是车牌识别系统中的关键环节。采用图像处理和神经网络相结合的方法设计新的车牌字符识别算法,先对分割出的车牌字符进行归一化处理,然后进行SOBEI．边缘检测和角点特征提取,最后输入BP神经网络进行训练、识别,其中BP神经网络模型属于改进型神经网络。通过一系列神经网络训练和仿真实验,车牌识别速度和正确率得到了明显的提高。     

特征级联在车辆检测中的应用与研究

对现有的采用机器学习算法检测车辆进行研究,分析其存在的不足;表现在特征或者算法单一,对光照等条件变化鲁棒性不够;针对这些问题,提出一种融合LBP特征与HOG特征,并结合Adaboost与SVM的车辆检测算法;借鉴级联的思想,首先采用AdaBoost对训练样本提取LBP特征进行训练,得到的分类器用于初步筛选并将其分类结果作为下一层分类器的输入;然后采用SVM算法对训练样本提取HOG特征进行训练,得到的分类器用于二次筛选上一层分类器的分类结果;实验结果证明Adaboost-SVM相结合的办法检测结果精度高,准确率和召回率均达到95%以上,FPPW与FPPI的值均在5%左右;同时由于算法采用的特征对光照条件具有较强的鲁棒性,因此光照条件的变化对算法的识别结果影响较低;实时性方面,每帧图像的处理时间为75 ms,满足实时性要求。     

Electron beam energy monitoring using thermoluminescent dosimeters and electron back scattering

Periodic checks of megavoltage electron beam quality are a fundamental requirement in ensuring accurate radiotherapy treatment delivery. In the present work, thermoluminescent dosimeters (TLDs) positioned on either side of a lead sheet at the surface of a water equivalent phantom were used to monitor electron beam quality using the electron backscattering method. TLD100 and TLD100H were evaluated as upstream detectors and TLD200, TLD400 and TLD500 were evaluated as downstream detectors. The evaluation assessed the test sensitivity and correlation, long and short term reproducibility, dose dependence and glow curve features. A prototype of an in-air jig suitable for use in postal TLD dose audits was also developed and an initial evaluation performed. The results indicate that the TLD100-TLD200 combination provides a sensitive and reproducible method to monitor electron beam quality. The light weight and easily fabricated in-air jig was found to produce acceptable results and has the potential to be used by radiation monitoring agencies to carry out TLD postal quality assurance audits, similar to audits presently being conducted for photon beams.     

基于OTSU分割和HOG特征的行人检测与跟踪方法

传统的HOG算法针对整幅图像进行行人特征提取,大量的非人窗口计算必然降低检测的准确率和效率。为此,提出一种基于OTSU分割和HOG特征的行人检测与跟踪方法。利用OTSU算法以最佳阈值分割图像,在分割区域的基础上进行Canny边缘检测,通过边缘的对称性计算确定行人候选区,继而采用经PCA方法降维后的HOG特征和隐马尔可夫模型对行人候选区进行检测验证。最后,以确定的行人区域为跟踪窗口,利用CamShift算法跟踪行人。多组实验结果证明,本文方法的行人检测效率和精度均有所提高,跟踪性能稳定、可靠。     

License plate location based on trichromatic imaging and color-discrete characteristic

License plate location (LPL) is the key part of automatic license plate recognition (LPR) system that plays an important role in many applications and a number of LPL methods based on color information have been proposed. However, some problems, such as country specific, similar color interference, lighting variation sensitive, time consuming, processing only one kind of LP with same color combination a time and omitting the case of more than one combinations in a LP, should be solved further. In this study, a color-based LPL method that consists of three modules: color edge extraction, denoising and searching, is proposed. Color edge extraction, the kernel of the proposed, is designed by color-discrete characteristics of license plates in the trichromatic wavebands. In the experiment, 1384 images taken from natural scenes in China and other 104 countries or regions are employed. Of which, 74 have been failed to locate the license plates. The success rate and average execution time are 94.7% and 57 ms, respectively.     

机械式冲床停上死点位置的控制策略设计与实现

车牌定位是车牌自动识别的第一步,而如何考虑当照影响是车牌定位是否成功的关键。本文提出了一种考虑不同光照条件的车牌定位方法,首先,针对不同光照条件下采集到的车牌图像明暗度的不同,利用多阈值处理方法得到车牌信息不丢失的、最佳的二值图像,然后,在其二值图像中首先利用灰度跳变定位车牌的上下边界,接着对字符垂直投影后的宽度进行统一的调整并以固定的字符间隔特征定位车牌的左右边界,从而完成车牌定位。最后,通过实验验证了该方法的有效性。     

联合多站阵元域数据的水下目标检测与跟踪

为了提高复杂海洋环境中目标的检测、跟踪性能,提出一种联合多站阵元域数据的水下目标检测与跟踪方法.该方法采用序列马尔科夫链蒙特卡洛思想对目标进行采样更新,通过对接收概率中的后验概率以及采样函数进行分解展开,并根据多站阵元域数据计算采样粒子的联合似然,在迭代过程中实现目标数目和目标状态的联合估计.研究结果表明,该方法对单目标的平均定位误差在较高信噪比下能够稳定在50 m以内,对多目标随机出入场景中新生及消失目标实现有效检测,同时对强干扰下弱目标及交叉目标实现有效检测跟踪。仿真结果和海试数据均验证该方法具有良好的目标检测与跟踪性能。      

液体可视图象序列中杂质小目标的检测方法（英）

在分析红外图象和可视图象差异的基础上,研究了一种适用于可视图象序列的运动小目标检测算法.受目标影响的象素点具有较弱的时间相关性,即相对普通象素点有较大的时间方差,根据此特征,用管道更新的方法产生新方差图象序列,再用管道更新的方法累积,增加信噪比,然后用基于统计均值的自适应门限方法分割出目标,最后将原图象的梯度倒数作为象素点强度的加权,有效地抑制灰度突变边界的传感器噪音,保存目标点.     

An alternative gamma tracking method using the Compton scattering probability

A new gamma tracking method, called probabilistic method and based on the Compton scattering probability of photons in a highly segmented Ge spectrometer, has been evaluated with the GEANT simulation code. This method has been used to calculate the performances of a 4 array composed of planar detectors as their crystal shape allows to realize identical segments of quasi-cubic shape. The tracking algorithm selects the full-absorption events by taking into account only the centre of gravity of the segments hit and the partial energies released in each of them. The results obtained, in terms of photopeak detection efficiency and peak-to-total ratio, are comparable to the ones deduced from the usual tracking methods. The probabilistic algorithm seems also to apply efficiently for other types of detectors as simulation calculations performed on a single stack of planar detector and a single segmented CLOVER detector give nearly the same reconstruction efficiency.Received: 26 November 2002, Revised: 29 April 2003, Published online: 23 March 2004PACS: 29.30.Kv X- and -ray spectroscopyG. Duchêne: Spokesperson     

An Improved Stereo Matching Algorithm for Vehicle Speed Measurement System Based on Spatial and Temporal Image Fusion

This paper proposes an improved stereo matching algorithm for vehicle speed measurement system based on spatial and temporal image fusion (STIF). Firstly, the matching point pairs in the license plate area with obviously abnormal distance to the camera are roughly removed according to the characteristic of license plate specification. Secondly, more mismatching point pairs are finely removed according to local neighborhood consistency constraint (LNCC). Thirdly, the optimum speed measurement point pairs are selected for successive stereo frame pairs by STIF of binocular stereo video, so that the 3D points corresponding to the matching point pairs for speed measurement in the successive stereo frame pairs are in the same position on the real vehicle, which can significantly improve the vehicle speed measurement accuracy. LNCC and STIF can be used not only for license plate, but also for vehicle logo, light, mirror etc. Experimental results demonstrate that the vehicle speed measurement system with the proposed LNCC+STIF stereo matching algorithm can significantly outperform the state-of-the-art system in accuracy.     

一种抗遮挡抗干扰的目标检测跟踪方法研究

研究了复杂背景环境下抗遮挡和抗干扰的目标检测跟踪技术。提出了基于轨迹预测关联的红外/可见光混合决策检测跟踪算法,开发了基于VPX总线架构的FPGA+DSP硬件模块和关键软件模块。结果表明,该方法综合利用红外与可见光两种图像的优势信息检测跟踪同一目标,能够有效地解决目标检测跟踪过程中抗遮挡和抗干扰的问题。     

Detection

This review on second- and third-generation multidetectors devoted to heavy-ion collisions aims to cover the last twenty years. The presented list of devices is not exhaustive but regroups most of the techniques used during this period for nuclear reactions at intermediate energy (≃ 10A MeV to 1A GeV), both for charged-particle and neutron detection. The main part will be devoted to 4π multidetectors, projectile decay fragmentation, high-resolution magnetic spectrometers, auxiliary detectors and neutron detection. The last part will present the progress in electronics and detection in view of the construction of future-generation detectors.     

Micro-crack detection of nonlinear Lamb wave propagation in three-dimensional plates with mixed-frequency excitation

We propose a nonlinear ultrasonic technique by using the mixed-frequency signals excited Lamb waves to conduct micro-crack detection in thin plate structures. Simulation models of three-dimensional(3D) aluminum plates and composite laminates are established by ABAQUS software, where the aluminum plate contains buried crack and composite laminates comprises cohesive element whose thickness is zero to simulate delamination damage. The interactions between the S_0 mode Lamb wave and the buried micro-cracks of various dimensions are simulated by using the finite element method.Fourier frequency spectrum analysis is applied to the received time domain signal and fundamental frequency amplitudes,and sum and difference frequencies are extracted and simulated. Simulation results indicate that nonlinear Lamb waves have different sensitivities to various crack sizes. There is a positive correlation among crack length, height, and sum and difference frequency amplitudes for an aluminum plate, with both amplitudes decreasing as crack thickness increased, i.e.,nonlinear effect weakens as the micro-crack becomes thicker. The amplitudes of sum and difference frequency are positively correlated with the length and width of the zero-thickness cohesive element in the composite laminates. Furthermore,amplitude ratio change is investigated and it can be used as an effective tool to detect inner defects in thin 3D plates.     

Consistent empirical physical formula construction for recoil energy distribution in HPGe detectors by using artificial neural networks

The gamma-ray tracking technique is a highly efficient detection method in experimental nuclear structure physics. On the basis of this method, two gamma-ray tracking arrays, AGATA in Europe and GRETA in the USA, are currently being tested. The interactions of neutrons in these detectors lead to an unwanted background in the gamma-ray spectra. Thus, the interaction points of neutrons in these detectors have to be determined in the gamma-ray tracking process in order to improve photo-peak efficiencies and peak-to-total ratios of the gamma-ray peaks. In this paper, the recoil energy distributions of germanium nuclei due to inelastic scatterings of 1–5 MeV neutrons were first obtained by simulation experiments. Secondly, as a novel approach, for these highly nonlinear detector responses of recoiling germanium nuclei, consistent empirical physical formulas (EPFs) were constructed by appropriate feedforward neural networks (LFNNs). The LFNN-EPFs are of explicit mathematical functional form. Therefore, the LFNN-EPFs can be used to derive further physical functions which could be potentially relevant for the determination of neutron interactions in gamma-ray tracking process.     

液体可视图象序列中杂质小目标的检测方法

在分析红外图象和视图象差异的基础上，研究了一种适用于可视图像序列的运动小目标检测算法。受目标影响的象素点具有较弱的时间相关性，即相对普通象素点有较大的时间方差，根据此特征，用管道更新的方法产生新方差图象序列，再用管道更新的方法累积，增加信噪比，然后用基于统计均值的自适应门限方法分割出目标，最后将原图象的梯度倒数作为象素点强度的加权，有效地抑制灰度突变边界的传感器噪音，保存目标点。     

Dose reassessment by using PTTL method in MTS-N (LiF:Mg,Ti) thermoluminescent detectors

The thermoluminescent (TLD) method is one of the most commonly used in dose measurements in radiation protection dosimetry. Due to its many advantages this method is widely spread. However, TLD has especially one disadvantage which is very inconvenient: the dose information in already read detectors is erased and in routine standard way the dose can not be reassessed. The positive is that this shortcoming can be eliminated by applying UV radiation. After first readout the same detector can be subjected to UV exposure and then read once again to reassess the dose.This method for reassessment of dose is based on phototransferred thermoluminescence (PTTL). In an irradiated TL detector deep traps are not emptied during the first readout. During exposure to UV, electrons are transferred from deep traps to shallower dosimetric traps. This TL signal emerging during the second readout following UV illumination is called phototransferred thermoluminescence.A method for reassessment of dose in a previously readout TLD is presented in this work. Experiments show that the method works well within region of doses between 5 and 50 mGy, but could be applied for higher doses as well. The efficiency of dose reassessment reaches about 17 percent of the first readout.The method could be a noticeable improvement in TLD dosimetry, giving more opportunities for better control and reliability of measurements.