Application of Gobor wavelet and SLLE in face recognition

In order to improve the recognition rate of face recognition algorithm, a new algorithm of face recognition is proposed based on Gabor wavelet transform and Supervised Locally Linear Embedding (SLLE). Gabor wavelet is introduced as a method to extract Gabor magnitude features by convolving the normalized face image with multi scale and multi orientation Gabor filters. In the feature extraction module, the dimension of Gabor features is reduced by SLLE. A minimum distance classifier is trained for classification. With the test of the ORL and YALE face database, it is found that 3.5 %~37.8% increase in recognition rate can be achieved compared with other algorithms.     

A face recognition method based on modular PCA and SVD

In order to solve the problem that modular PCA method is sensitive to translation, rotation and other geometric transform, a face recognition method based on modular PCA and singular value decomposition (SVD) is proposed. The PCA features of sub image and SVD features are extracted respectively. The distance measure that fuses information of modular PCA and SVD is obtained. Minimum distance classifier is used to face recognition. Experimental results on ORL human face database show that the proposed method can obtain higher recognition rate.     

光强倒数色度空间的彩色人脸光照预处理

为了解决彩色人脸识别中色彩信息易受光照影响的问题,提出一种基于光强倒数色度空间(IICS)的彩色人脸图像预处理方法。本方法首先将图像均匀地分割成子块;将每个图像块变换为IICS空间中的一个二维数据集,并根据数据集的线性分布特性估计图像块的光照颜色;然后对全部图像块的光照估计进行颜色直方图统计,根据直方图对分块估计的结果进行合并;最后,利用估计得到的光照和对角模型将图像光照校正到标准白光下,用于人脸识别。在AR和FERET人脸库上的实验表明,通过引入本光照预处理,有效增强了彩色人脸识别方法对光照变化的鲁棒性,提高了识别精度。     

Face recognition method with local matching

From human cognition, a face recognition method with local matching based on statistical learning is proposed. The image is divided into several subimages and each subimage is considered as a weak classifier. The Adaboost learning algorithm is used to train the weak classifiers and construct a strong classifier. As a result, all subimages are effectively combined together to explore the best discriminating power and improve the classification accuracy. Compared with the holistic matching methods, the local matching method is robust to variations in illumination, expression, and pose, etc. The experimental results show that the proposed method can improve the face recognition accuracy and is robust to variations in illumination and expression.     

基于粒子群算法和支持向量机的黄花菜叶部病害识别

使用数字图像处理技术,以黄花菜叶部病害图像为识别对象,基于Lab空间和K-means聚类算法分割病害区域,提取目标区域的颜色特征、方向梯度直方图(histogram of oriented gradient,HOG)特征和形状特征,分别建立单一特征模型和特征融合模型,采用粒子群(particle swarm optimization,PSO)算法通过交叉验证优化支持向量机(support vector machine,SVM)模型的惩罚因子和核参数,建立基于PSO-SVM的多特征融合分类模型识别黄花菜病害。基于SVM的多特征融合分类模型识别率高于单一特征分类模型,识别率可达为81.67%;基于PSO-SVM多特征融合分类模型识别率高达92.39%。基于PSO-SVM的多特征分类模型识别率高,可以及时、便捷、高效地识别黄花菜病害。     

Automatic head recognition by integrating mean shift with multiple features

To improve the head detection accuracy in video sequences captured with fixed vertical monocular camera, a novel method of head recognition based on mean shift and multiple features is proposed. Firstly, mean shift based image segmentation algorithm with color information and spatial information is suggested to derive the candidate head components in images. Furthermore, by referring to two features that the contour of human head regions are approximate round and the hair color distribution is clustered, the evaluation models based on the contour information and inside color information of candidate head components are presented for head recognition. The experimental results show that the proposed algorithm can effectively reduce the light interfere and eliminate fake target whose color information is similar to hair color distribution. The detection rate for static images can reach about 89.4%.     

Research and implementation on handwritten numeralrecognition technique based on fuzzy mode

Based on simulation and numeral recognition of the 24-bit bmp images of postal codes obtained by CCD transducer, a binarization method based on coloring matter is proposed, which is employed to extract the binary image of the frame line of postal code and the binary image of the mail character separately. The procedure of image processing is analyzed, which includes detecting and separating borderlines of postal code frames, image binarization, smoothing noise-removal, slant rectification, extraction of postal code numerals, position normalization processing, subdivision of digital image, feature extraction of character recognition, and so on. The pretreatment process of image recognition is described. Then, character feature extraction is conducted for pretreated images, while the angle correctness for postal code frames and characters is performed. Finally, the recognition result is determined. In this method, the difficulty of separating frame line from postal code character line is eliminated. In order to improve recognition rate, the combination of many methods of handwritten numeral is used, resulting in improved recognition rate of postal codes. The experiment shows that this method can obtain higher recognition rate, providing reference for the separation technique in postal system.     

基于最小阈值法的棉花幼苗识别研究

 面对纯机械式分苗机构经常出现的漏分和多分的问题,本文提出使用基于最小阈值法的图像识别技术对棉花幼苗进行识别和分离。使用了HSV颜色空间内与光照条件无关的H值将RGB真彩色图像转化为灰度图像,以提高图像处理的实时性。基于最小值阈值法,提出一种能够有效地将处于复杂背景下的棉花幼苗茎部图像提取的方法,使用棉花幼苗茎部图像的面积和长宽比等几何信息能有效识别棉花幼苗,并得到棉花幼苗茎部的重心坐标。本文的研究证明了用图像处理中的最小阈值法识别棉花幼苗是可行的。     

Face Recognition Based on Weighted PCA

This paper proposes a face novel recognition method based on Weighted PCA, which combined weighted features with PCA. At first, we calculate the weighted subspace for each class by minimizing the weighted reconstruction error. Then a test example is classified by the distance from weighted subspace. The experiments on Cambridge ORL database show that our Weighted PCA method can improve the recognition rate significantly without increasing the computation, when compared with PCA.     

Affine invariant shape recognition with particle swarmoptimization algorithm

Focusing on the problem that affine transformation will exist among the contour images due to variation of the viewpoints, a new approach to extract affine invariant features and matching strategy is proposed for shape recognition. First, the centroid distance and azimuth angle of each boundary point are computed. Then, with a prior defined angle interval, all the points in the neighbor region of the sample point are considered to calculate the average distance for eliminating noise. After that, the centroid distance ratios(CDRs) of any two contour points with angle difference of 180° are achieved as the representation of the shape, which would be invariant to affine transformation. Since the angles of contour points changed non linearly among affine related images, the CDRs should be resampled to build corresponding relationship. It could be regarded as an optimization problem of path planning. In our method, a PSO based path planning model is presented to address this problem. The experimental results demonstrate the efficiency of the proposed method in shape recognition with translation, scaling, rotation, distortion and noise interference.     

频域滤波的水泥路面图像降噪增强方法

针对在水泥路面裂缝图像识别中刻槽和纵向条带状形成的噪声干扰,提出了一种基于频域滤波的降噪增强方法。首先,分析了这两类噪声的频谱特征,并根据空域图像中刻槽的周期性,推导了频域图像中刻槽谱峰的位置,设计了消除这些谱峰的系列滤波器。根据纵向条带状干扰属于低频成分的特性,设计了抑制低频、增强高频的滤波器,经优化滤波器参数达到消除这类干扰的效果。实验结果表明,该方法能够消除两类噪声干扰,同时,增强了裂缝的对比度,为水泥路面裂缝识别提供了理论基础。     

Recognition Algorithm of Motorway Curl Road

Road detection is a sub-system of vehicle intelligent auxiliary drive system, which includes such sub-projects as road segmentation and road recognition. The image from CCD camera is processed, and road line is abstracted from it through image process. Then the direction of the road, such as left or right, can be identified according to the road line. Because of the special requirement, i.e. the rigid real-time requirement, the algorithm adopted must have the very low time cost. Through many experiments, a kind of recognition algorithm is proposed based on region-growing process and curve fitting, which was verified to have good effect.     

Study on the Recognition of Transformer Winding Deformation Using Wavelet Transform

This paper illustrates the methods of extracting characteristic parameters and deformation recognition using wavelet transfor,which applied in low voltage impulse(LVI) method on the transformer winding deformation detection.The results show that it is effective to eliminate the noise using wavelet transform,meanwhile,it can also make measurement has repetitivity and has a property on model recognition on defomation.     

Recognition and prediction of fracture by using antcolony algorithm analysis in carbonate reservoir

A fracture-developed carbonate oil field in the east has a complicate oil and gas distribution in facture network at late development stage, which leads to multiplicity of favorable reservoir estimation. In view of problems above, ant colony algorithm is adapted to recognition and predict facture in this field. The ant parameters are optimized on the basis of core and image log data, and the spatial distribution feature of fracture is described, as a reticulate structure with three dominant clusters of fracture (NW, NNE & NE). Drilling leakage, production data and image log are then used for reliability analysis of fracture predict, which presents that ant colony algorithm is a practicable methodology to recognition fracture and provides a support for remaining oil distribution analysis. As a booming bionic algorithm, ant colony algorithm has great potential for quantitative fracture predict with seismic materials.     

Image edge detection algorithm in fuzzy domain

An image edge detection algorithm in fuzzy domain is proposed, which combines adaptive fuzzy enhancement and multi direction fuzzy morphology to detect the edges of fuzzy image. The adaptive fuzzy enhancement method enhances the fuzzy image within blocks with sliding windows to avoid losses of the real edges resulting from enhancing with single threshold for the whole image and lead strong adaptive ability to image region variance. The multi direction fuzzy mathematical morphology operates on the enhanced fuzzy image with structure elements of multiple different directions to extract the real edges with directionality and restrain non directional noise. Experiments show the algorithm can detect fuzzy image edge effectively with strong antinoise ability.     

Human Face Location Based on Motion Information in a Complex Background

This paper proposes an improvement for the traditional Optical Algorithm, and presents a new way to image segmentation in a complex background. In addition, combined with the neural network, the system can locate the possible human faces successfully by means of two-step location model. In our system, the searching and locating of the human face is the most important stage. According to this, the authors adopt the two-step way to run, firstly they take up the segmentation of the candidate human face areas and then the accurate face locating based on the neural network is used. This algorithm is fast and robust. Experimental results with real scene images are given out there, and all these prove that two-step method gains many advantages in the course of human face location with motion information, such as real-time, robustness and practicality. In addition, the proposed system is also the fundamental and important part of the perfect human face recognition system.     

基于RGB线性组合模型的柑橘果实为害状识别

为了准确识别柑橘果实病虫害,提高柑橘生产信息化水平,本研究提出了最优RGB线性组合颜色模型来进行目标识别。首先在具有代表性的几幅图像中选取为害状区域与正常区域的点若干,统计这些样本点的R、G、B值及其均值,设计为害状区域与正常区域灰度差最大线性规划目标函数并求解,建立识别模型。最后利用识别模型结合阈值分割法对采集的96幅柑橘果实图像进行处理,发现识别正确率、误检率、漏检率分别81.25%,14.58%,4.17%,识别效果良好。与2R-G-B模型,G-B模型和R-B模型相比,本方法为害状区域与正常区域灰度差最大,故识别正确率最高,误检率、漏检率最低。试验结果表明：RGB线性组合模型法可用于柑橘果实病虫害的识别。     

多邻域保持嵌入的人脸识别方法

现有流形学习算法在学习人脸数据时,假设所有数据点位于单一低维嵌入流形之上,当数据点实际分布在不同的流形上时,单流形假设就会影响数据真实空间结构。为此提出一种基于多邻域保持嵌入（multiple neighborhood preserving embedding, M-NPE）的学习算法来发现不同类别数据在不同维度的低维嵌入空间中分布的多流形结构。首先,单独学习不同类别数据的流形,得到反映其本质特征的流形;再通过遗传算法搜索每个流形的最优维数;最后依据最小重构误差分类器对样本分类。在Extended Yale B和CMU PIE这2个大型人脸库上实验结果验证了该算法的有效性。     

基于改进SVM分类器的动作识别方法

利用智能手机加速度传感器信号,提出一种改进的动作识别方法以降低传统动作识别方法的复杂程度,提高识别率。在特征提取时用盲选法,即用PCA(principal component analysis)进行特征值的降维和去除多维间的干扰,而所选特征没有对应的物理意义;并在分类识别中将遗传算法应用到SVM(support vector machine)分类器参数优化中。通过实验表明,该方法能够对日常的走路、站立、跑及上下楼等动作进行准确的识别。     

基于迁移学习的园艺作物叶部病害识别及应用

为给农户提供质优价廉的园艺作物叶部病害识别服务,提出基于迁移学习的模型训练及基于Flask的Web部署方法。对PlantVillage数据集进行预处理,分别使用ResNet18、ResNet50和ResNet152 3种模型进行迁移学习训练,得到3种识别模型。利用Flask将模型部署到服务器上。3种识别模型对苹果等14类园艺作物26种叶部病害的平均识别准确率分别是95.61%、96.63%和97.33%,识别单张图像的时间分别是10.9、17.9、33.7 ms。综合考虑模型特点和用户期望,设计快速、标准和准确3种识别模式,实现深度模型在服务器中稳定运行,具有一定的实用价值。