基于多核的车牌识别的架构实现

多核技术是现在提高芯片性能的主要方法。区别于传统以PC和DSP为核心的车牌识别系统,以FPGA为核心,利用SOPC技术构建了车牌识别多核处理器。给出了一种基于多核的车牌识别架构,在该多核处理器中,以3个Nios II 软核为主要处理器核处理车牌定位、字符特征识别提取及识别等处理,同时构建硬件加速器作为协处理器处理图像增强、边缘检测和膨胀、腐蚀等数学形态学处理。在CQ片上路由器基础上,构建了NOC用以实现片上多核通信。另外,为了保证路由器与多处理器核之间的快速、并行通信,加入了数据驱动模块。整个系统在Altera Cyclone IV FPGA上实现了车牌的识别。这种片上系统设计方法具有硬件设计灵活,可扩展性强等优点,能有效地降低系统软硬件设计的难度,缩短开发周期,并提高设计的可靠性。     

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.     

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。     

基于实时样本采集的个性化手写汉字输入系统设计

手写汉字识别是手写汉字输入的基础。目前智能设备中的手写汉字输入法无法根据用户的汉字书写习惯,动态调整识别模型以提升手写汉字的正确识别率。通过对最新深度学习算法及训练模型的研究,提出了一种基于用户手写汉字样本实时采集的个性化手写汉字输入系统的设计方法。该方法将采集用户的手写汉字作为增量样本,通过对服务器端训练生成的手写汉字识别模型的再次训练,使识别模型能够更好地适应该用户的书写习惯,提升手写汉字输入系统的识别率。最后,在该理论方法的基础上,结合新设计的深度残差网络,进行了手写汉字识别的对比实验。实验结果显示,通过引入实时采集样本的再次训练,手写汉字识别模型的识别率有较大幅度的提升,能够更有效的满足用户在智能设备端对手写汉字输入系统的使用需求。     

基于深度残差网络的脱机手写汉字识别研究

摘 要: 手写汉字识别是模式识别与机器学习的重要研究方向和应用领域。近年来,随着深度学习理论方法的完善、新技术的层出不穷,深度神经网络在图像识别分类、图像生成等典型应用中取得了突破性的进展,其中,深度残差网络作为最新的研究成果,已成功应用于手写数字识别、图片识别分类等多个领域。本文将研究深度残差网络在脱机孤立手写汉字识别中的应用方法,通过改进残差学习模块的单元结构,优化深度残差网络性能,同时通过对训练集的预处理,从数据层面实现训练生成模型性能的提升,最后设计实验,验证深度残差网络、End-to-End模式在脱机手写汉字识别中的可行性,分析、总结存在的问题及今后的研究方向。     

Spin-image surface matching based target recognition in laser radar range imagery

We explore the problem of in-plane rotation-invariance existing in the vertical detection of laser radar (Ladar) using the algorithm of spin-image surface matching. The method used to recognize the target in the range imagery of Ladar is time-consuming, owing to its complicated procedure, which violates the requirement of real-time target recognition in practical applications. To simplify the troublesome procedures, we improve the spin-image algorithm by introducing a statistical correlated coefficient into target recognition in range imagery of Ladar. The system performance is demonstrated on sixteen simulated noise range images with targets rotated through an arbitrary angle in plane. A high efficiency and an acceptable recognition rate obtained herein testify the validity of the improved algorithm for practical applications. The proposed algorithm not only solves the problem of in-plane rotation-invariance rationally, but also meets the real-time requirement. This paper ends with a comparison of the proposed method and the previous one.     

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。     

An Orientation‐Independent Imaging Technique for the Classification of Blood Cells

This study presents an orientation independent imaging technique for the classification and recognition of blood cells with relevant applications to related problems in cytometry and medical diagnosis. The proposed method integrates three important aspects towards its practical implementation: 1. The use of the principal component (PC) transform to reorient in an optimal fashion the image data and to make of the feature matching an orientation independent process; 2. The establishment of similarity measures to quantify the matching process of the shape of the blood cell populations with any chosen degree of certainty; and 3. The application of the recognition and classification processes once all the similarity measurements are gathered. An extension of this two‐dimensional (2‐D) method to a three‐dimensional (3‐D) base is proposed. Results of this technique using real‐world image data of blood cell populations are given. It is appropriate to note that although the images used here are real‐world samples of blood cell populations, other samples containing different biological specimens could have been used just as well.     

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.     

细胞显微图像的处理与识别系统研究

介绍了模板匹配和特征识别两种识别方案,采用数字图像处理技术和模式识别技术,实现了细胞显微图像中的细胞自动识别与分类.实验结果表明,特征识别具有直接性、运算量小等特点,为细胞显微图像的模式识别提供了可靠有效的方法.     

基于改进ViT的红外人体图像步态识别方法研究

针对卷积神经网络在步态识别时准确率易饱和现象,以及Vision Transformer(ViT)对步态数据集拟合效率较低的问题,提出构建一个对称双重注意力机制模型,保留行走姿态的时间顺序,用若干独立特征子空间有针对性地拟合步态图像块;同时,采用对称架构的方式,增强注意力模块在拟合步态特征时的作用,并利用异类迁移学习进一步提升特征拟合效率。将该模型运用在中科院CASIA C红外人体步态库中进行多次仿真实验,平均识别准确率达到96.8%。结果表明,本文模型在稳定性、数据拟合速度以及识别准确率3方面皆优于传统ViT模型和CNN对比模型。     

Matching a transducer to water at cavitation: acoustic horn design principles

High-power ultrasound for several decades has been an integral part of many industrial processes conducted in aqueous solutions. Maximizing the transfer efficiency of the acoustic energy between electromechanical transducers and water at cavitation is crucial when designing industrial ultrasonic reactors with large active volumes. This can be achieved by matching the acoustic impedances of transducers to water at cavitation using appropriately designed ultrasonic horns. In the present work, a set of criteria characterizing the matching capabilities of ultrasonic horns is developed. It is shown that none of the commonly used tapered-shape horns can achieve the necessary conditions. An analytical method for designing five-element acoustic horns with the desirable matching properties is introduced, and five novel types of such horns, most suitable for practical applications, are proposed. An evaluation of the horns' performance is presented in a set of experiments, demonstrating the validity of the developed theoretical methodology. Power transfer efficiency increase by almost an order of magnitude is shown to be possible with the presented horn designs, as compared to those traditionally utilized.     

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.     

伪Zernike矩在低信噪比红外目标检测中的应用

由于探测器本身固有的特性以及军事上的极限使用要求,红外图像普遍存在目标-背景间对比度较差、目标边缘模糊和噪声较大等特点,采用常规模板匹配、Hu矩方法难以取得理想的检测效果。针对低信噪比红外目标图像,分析了伪Zernike矩的基本原理、不变性和计算方法,提出了低信噪比红外目标检测的伪Zernike矩方法,并比较了模板匹配、Hu矩、伪Zernike矩方法的目标检测效果。理论分析和实验验证了该方法的有效性。     

一种改进的块匹配运动目标识别算法

鉴于传统的块匹配法在图像特征提取、匹配以及搜索等方面有诸多不足,复杂的边缘提取处理与基于灰度的匹配准则都使得系统的处理速度与精度降低,为了使识别处理更加快速精确,提出了一种改进的块匹配识别算法。通过拉普拉斯算子对图像特征进行增强,提高图像的特征信息;实验表明,针对320像素240像素大小的图像,改进算法的特征增强计算时间在Matlab中平均约为0005 4 s。     

强度不变的自动模式识别技术

针对光学模式识别技术用化要求，提出一种新的图像预处理算法，使识别现场光照明强度及分布的变化对识别效率所带来的影响大大降低。采用该算法进行实时硬件光学相关试验结果证实应用该算法可以实现强度的自动目标识别。     

Fuzzy Matching Based on Gray-scale Difference for Quantum Images

Quantum image processing has recently emerged as an essential problem in practical tasks, e.g. real-time image matching. Previous studies have shown that the superposition and entanglement of quantum can greatly improve the efficiency of complex image processing. In this paper, a fuzzy quantum image matching scheme based on gray-scale difference is proposed to find out the target region in a reference image, which is very similar to the template image. Firstly, we employ the proposed enhanced quantum representation (NEQR) to store digital images. Then some certain quantum operations are used to evaluate the gray-scale difference between two quantum images by thresholding. If all of the obtained gray-scale differences are not greater than the threshold value, it indicates a successful fuzzy matching of quantum images. Theoretical analysis and experiments show that the proposed scheme performs fuzzy matching at a low cost and also enables exponentially significant speedup via quantum parallel computation.     

A novel pose and illumination robust face recognition with a single training image per person algorithm

In the real-world application of face recognition system, owing to the difficulties of collecting samples or storage space of systems, only one sample image per person is stored in the system, which is so-called one sample per person problem. Moreover, pose and illumination have impact on recognition performance. We propose a novel pose and illumination robust algorithm for face recognition with a single training image per person to solve the above limitations. Experimental results show that the proposed algorithm is an efficient and practical approach for face recognition.     

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.