Description: Support Vector Machine Classifier(SVM)
基于支持向量基的图像分类方法，程序变得很好，国外网站上下的，是一个完整的工程-Support Vector Machine Classifier (SVM) based on support vector-based image classification methods, procedures have become very good, abroad site from top to bottom, is a complete project Platform: |
Size: 8192 |
Author:haoyubao |
Hits:
Description: 目的:运用强化学习!多分类器集成!降维方法等最新计算机技术,结合细胞病理知识,设计制作/智能化肺癌细胞病理图像诊断系统0"方法:采集细胞图像,运用基于强化学习的图像分割法将细胞区域从背景中分离出来 运用基于样条和改进2方法对重叠细胞进行分离和重构 提取40个细胞特征用于贝叶斯!支持向量机!紧邻和决策树4种分类器,集成产生肺癌细胞分类结果 建立肺癌细胞病理图库,运用基于等降维方法对细胞进行比对,给予未定型癌细胞分类"结果:/智能化肺癌细胞病理诊断系统0应用于临床随机1200例肺部病灶穿刺细胞学涂片,肺癌识别诊断率94180 ,假阳性率1185 ,假阴性率3135 ,肺癌分类识别率82190 ,核异型细胞识别率74120 "结论:/智能化肺癌早期细胞病理诊断系统0对肺癌细胞涂片诊断率高,克服了肺癌细胞病理诊断过程中取检细胞数量少,重叠细胞识别率低,涂片背景及染色差异等干扰因素,可辅助临床肺部病灶的穿刺细胞病理诊断"-Objective Design and develop a intelligent cytopathological lung cancer diagnosing system(ICLCDS) utilizing the latest computer technologies(including Reinforcement Lcaming Multiple Classifier Fusion and Dimcnsionality Reduction) and the cy-topathological knowledge on lung canccrcclls Methods We got information ofcclls and segregated cell regions in a slice image using an magi scgmcntouon a址orithm Sascd on reinforcement lcaming including rcconstmction of overlapped cell area Sascd on B一Spline and improved dcBoor-Cox Mcthoc} We comSincd multiple classifiers including Baycsian classific:Support Vector Machine(SVM)
classific K-Ncarcst NcighSour( KNN) and Decision c classific to achieve an accurate result of cytopathological lung cancer diag-nosis Results Experimental results on 1 200 cases randomly selected we as follows the accurate diagnosis rate for lung cancer idcn-tification was the false positive rate was 1. 8`J /c‘the false negative rate was 3. 3`J /c‘the type class Platform: |
Size: 221184 |
Author:高阳 |
Hits:
Description: Abstract
We present a component-based, trainable system for detecting
frontal and near-frontal views of faces in still gray
images. The system consists of a two-level hierarchy of Support
Vector Machine (SVM) classifiers. On the first level,
component classifiers independently detect components of
a face. On the second level, a single classifier checks if the
geometrical configuration of the detected components in the
image matches a geometrical model of a face. We propose
a method for automatically learning components by using
3-D head models. This approach has the advantage that
no manual interaction is required for choosing and extracting
components. Experiments show that the componentbased
system is significantly more robust against rotations
in depth than a comparable system trained on whole face
patterns. Platform: |
Size: 349184 |
Author:a |
Hits:
Description: 此方法采用经典的PCA对人脸图像进行特征提取，用libsvm库函数的SVM分类器对图像分类。-This method uses the classical PCA on the face image feature extraction, with the libsvm library function of SVM classifier for image classification. Platform: |
Size: 6144 |
Author:zhangpei |
Hits:
Description: 利用级联SVM的人体检测方法从图像中检测出人体是计算机视觉应用中的关键步骤。通过一个由简到繁的级联线性SVM分类器将级
联拒绝的机制与梯度方向直方图特征相结合,实现了一个准确和快速的人体检测器,整个检测器由级联的线性
SVM分类器组成。实验结果表明,在保持Dalal算法检测准确性的同时,大幅的提高了检测速度,每秒平均可以处
理12帧左右的320 ×240的图像。-Human detection using cascade SVM method detected from the images of computer vision applications, the human body is a key step. From the simple to the complex through a cascade of linear SVM classifier with the cascade mechanism of rejection combined histograms of oriented gradients to achieve an accurate and rapid detection of the human body, the detector by a cascade of linear SVM classifier component. The results show that the accuracy in maintaining Dalal algorithm to detect the same time, substantially increase the detection rate of 12 frames per second on average can process about 320 × 240 image. Platform: |
Size: 482304 |
Author:lilin |
Hits:
Description: 一篇很不错的关于人脸表情识别的论文。论文提出了一种基于人脸局部特征的表情识别方法，先选取人脸重要的局部特征,对得到的局部特征进行主成分分析,然后用支持向量机( SVM)设计局部特征分类器来确定测试表情图像中局部特征,同时设计支持向量机( SVM)表情分类器,确定表情图像的所属类别。-A very good facial expression recognition on paper. This paper proposes a feature based on local expression of face recognition, face first select the important local features, local features of the obtained principal component analysis, and support vector machine (SVM) classifier design to determine the local characteristics of test expression image local features, while design of support vector machine (SVM) classifier expression, determine the expression of the image category. Platform: |
Size: 428032 |
Author:王二 |
Hits:
Description: 基于SVM分类器的三域特征融合图像隐写分析算法。三域分别是指dct域，dwt域和空域。-A Steganalysis Based on the Fusion of Three-domain Features of Image and SVM (Support Vector Machines) Classifier.the three domain is the dct,the dwt and the space Platform: |
Size: 169984 |
Author:rita |
Hits:
Description: 支持向量机(svM)是一种新的机器学习技术。本文采用一对一方法构建多分类SVM
分类器。利用常用的灰度共生矩阵方法提取图像纹理特征，组成特征向量，输入构建好的SVM
多分类器中进行分类。对从Brodatz纹理库中选取的4张纹理图像进行了分类实验，取得较好的
分类结果-Support vector machine (svM) is a new machine learning techniques. In this paper, one way to build a multi-classification SVM classifier. GLCM using methods commonly used to extract image texture features, compositions of the vector input to build a good classifier in the SVM multi-classification. From the Brodatz texture library texture selected four images were classified experiments to obtain better classification results Platform: |
Size: 375808 |
Author:刘东 |
Hits:
Description: 的研究彩色数字图像的计算机分类识别方法并应用于古瓷片的自动分类。方法提出
了一种色彩纹理特征的提取模型，采用该模型，利用IGabor滤波器提取数字图像的色彩纹理特征，
并构造支持向量分类机(SVM)分类器组。结果实现了高准确率多类别图像的自动分类识别，并
成功应用于古瓷片的自动分类。结论色彩纹理特征提取方法将颜色与纹理进行融合，增强了数
字图像之间的特征区分能力。-Study color digital image classification and recognition method and a computer chip used in the automatic classification of porcelain. Method presents a color texture feature extraction model, using the model, the use of IGabor filter extracts the color digital image texture features, and construct support vector classification machine (SVM) classifier group. Results to achieve a high accuracy automatic multi-class image classification and recognition, and successfully applied to automatic classification of porcelain pieces. Conclusion color texture feature extraction method for integration of color and texture to enhance the digital image characteristics between the ability to distinguish between. Platform: |
Size: 346112 |
Author:刘东 |
Hits:
Description: 提出一种多尺度方向(multi-scale orientation,简称 MSO)特征描述子用于静态图片中的人体目标检
测.MSO 特征由随机采样的图像方块组成,包含了粗特征集合与精特征集合.其中,粗特征是图像块的方向,而精特征
由 Gabor 小波幅值响应竞争获得.对于两种特征,分别采用贪心算法进行选择,并使用级联 Adaboost 算法及 SVM 训
练检测模型.基于粗特征的 Adaboost 分类器能够保证高的检测速度,而基于精特征的 SVM 分类器则保证了检测精
度.另外,通过 MSO 特征块的平移,使得所提算法能够检测多视角的人体.通过对于 MSO 特征块的装配,使得算法能
够检测人群中相互遮挡的人体目标.在INRIA公共测试集合及SDL多视角测试集合上的实验结果表明,算法具有对视角与遮挡的鲁棒性和较高的检测速度. -The multi-scale orientation (MSO) features for pedestrian detection in still images are put forwarded in
this paper. Extracted on randomly sampled square image blocks (units), MSO features are made up of coarse and
fine features, which are calculated with a unit gradient and the Gabor wavelet magnitudes respectively. Greedy
methods are employed respectively to select the features. Furthermore, the selected features are inputted into a
cascade classifier with Adaboost and SVM for classification. In addition, the spatial location of MSO units can be
shifted, are used to the handle multi-view problem and assembled therefore, the occluded features are completed
with average features of training positives, given an occlusion model, which enable the proposed approach to work
in crowd scenes. Experimental results on INRIA testset and SDL multi-view testset report the state-of-arts results on
INRIA include it is 12.4 times the faster than SVM+HOG method. Platform: |
Size: 1868800 |
Author:尹世荣 |
Hits:
Description: 基于神经网络的遥感图像分类取得了较好的效果,但存在固有的过学习、易陷入局部极小等缺点.支持向量机机器学习方法,根据结构风险最小化(SRM)原理,表现出很多优于其他传统方法的性能,本研究的基于多类支持向量机分类器的遥感图像分类取得了达95.4 的分类精度.但由于遥感图像分类类别多,所需训练样本较大,人工选择效率较低,为此提出以人工选择初始聚类质心、C均值模糊聚类算法自动标注训练样本的基于多类支持向量机的半监督式遥感图像分类方法,期望能在获得适用的分类精度的基础上有效提高分类效率-Neural net based remote sensing image classification has obtained good results. But neural net has inherent
flaws such as overfitting and local minimums. Support vector machine (SVM), which is based on Structural Risk Min-
imization(SRM), has shown much better performance than most other existing machine learning methods. Using mul-
ti-class SVM classifier high class rate of 95.4 is obtained. But for the class number of remote sensing image is much
great, manually obtaining of training samples is a much time-consuming work. So a multi-class SVM based semi-super-
vised approach is presented. It is choosed that the initial clustering centroids manually first, then label the samples as
the training ones automatically with fuzzy clustering algorithm. It is believed that this method will upgrade the classifi-
cation efficiency greatly with practicable class rate Platform: |
Size: 25600 |
Author:cissy |
Hits:
Description: 基于SVM支持向量机的图像前景和背景分割代码他，通过手动选择训练样本生成分类器，然后进行分割-SVM SVM based image segmentation code foreground and background, he generated classifier by manually selecting training samples, and then split Platform: |
Size: 14336 |
Author:王莉 |
Hits:
Description: 一个自己封装的SVM类，可以实现对正负样品图片的剪切，分类器的训练，利用原来的文件测试分类器，利用一段视频测试分类器，同时也可以完成分类器的引用，同时SVM的训练参数可调。-Reference to a category of their own package of SVM, can achieve the image on the positive and negative samples of shear, training a classifier, using the original file test the classifier, using a video test classifier. At the same time, it can complete the classifier and SVM training parameters adjustable. Platform: |
Size: 24354816 |
Author:黄志举 |
Hits:
Description: 本示例展示了怎样从一个预处理的卷积神经网络中提取特征，并用这些特征去训练一个图像分类器。(This example shows how to extract learned features from a pretrained convolutional neural network, and use those features to train an image classifier. Feature extraction is the easiest and fastest way use the representational power of pretrained deep networks. For example, you can train a support vector machine (SVM) using |fitcecoc| (Statistics and Machine Learning Toolbox(TM)) on the extracted features.) Platform: |
Size: 372736 |
Author:RockLiu
|
Hits:
Description: 本文件是图像场景识别并进行分类的程序，已运行成功。
分别利用1 tiny image描述和最近邻分类器
2 bags of sifts描述和最近邻分类器
3bags of sifts描述和线性svm分类器进行场景分类识别的。
在主程序proj3中将FEATURE 改成tiny image，CLASSIFIER 改成nearest neighbor，注释其他FEATURE 和CLASSIFIER的选择就可以实现第一种场景分类识别：tiny image描述和最近邻分类器。以此类推就可以依次改FEATURE 和CLASSIFIER 形成第二种情况：bags of sifts描述和最近邻分类器和第三种情况：bags of sifts描述和线性svm分类器进行场景分类识别。(This document is an image scene recognition and classification of procedures have been run successfully.) Platform: |
Size: 18185216 |
Author:wujie123 |
Hits:
Search - SVM image classifier