Face recognition using decimated redundant discrete wavelet transforms

As discrete wavelet transform (DWT) is sensitive to the translation/shift of input signals, its effectiveness could be lessened for face recognition, particularly when the face images are translated. To alleviate drawbacks resulted from this translation effect, we propose a decimated redundant DWT (DRDWT)-based face recognition method, where the decimation-based DWTs are performed on the original signal and its 1-stepshift, respectively. Even though the DRDWT realizes the decimation, it enables us to explore the translation invariant DWT representation for the periodic shifts of the probe image that is the most similar to the gallery images. Therefore, it can solve the problem of translation sensitivity of the original DWT and address the translation effect occurring between the probe image and the gallery image. To further improve the recognition performance, we combine the global wavelet features obtained from the entire face and the local wavelet features obtained from face patches to represent both holistic and detail facial features, apply separate classifiers to global and local features and combine the resulted global and local classifiers to form an ensemble classifier. Experimental results reported for the FERET and FRGCv2.0 databases show the effectiveness of the DRDWT method and quantify its performance.     

基于卷积神经网络的遥感图像分类研究

遥感图像分类是模式识别技术在遥感领域的具体应用,针对遥感图像处理中的分类问题,提出了一种基于卷积神经网络(convolutional neural networks,CNN)的遥感图像分类方法,并针对单源特征无法提供有效信息的问题,设计了一种多源多特征融合的方法,将遥感图像的光谱特征、纹理特征、空间结构特征等按空间维度以向量或矩阵的形式进行有效融合,以此训练CNN模型。实验表明,多源多特征相融合能够加快模型收敛速度,有效提高遥感图像的分类精度;与其他分类方法相比,CNN能够取得更高的分类精度,获得更优的分类效果。     

A sliding window technique for efficient license plate localization based on discrete wavelet transform

Real-time license plate recognition (LPR) is an interesting but complicated research topic. Some previous works use discrete wavelet transform (DWT) to extract license plate (LP), however, most of them are not capable of dealing with complex environments such as the low-contrast source images and the dynamic-range problems. In this paper, we propose an algorithm for license plate localization (LPL) based on DWT. The LP can be extracted from different quality of source images under complex environments by using two frequency subbands. We first use the HL subband to search the features of LP and then verify the features by checking whether in the LH subband there exists a horizontal line around the feature or not. The proposed method can extract both front and back LPs of various vehicles. The experimental results show that the proposed method can achieve good LPL results with both short run-time and high accurate detection rate.     

Invariant pattern recognition using contourlets and AdaBoost

In this paper, we propose new methods for palmprint classification and handwritten numeral recognition by using the contourlet features. The contourlet transform is a new two dimensional extension of the wavelet transform using multiscale and directional filter banks. It can effectively capture smooth contours that are the dominant features in palmprint images and handwritten numeral images. AdaBoost is used as a classifier in the experiments. Experimental results show that the contourlet features are very stable features for invariant palmprint classification and handwritten numeral recognition, and better classification rates are reported when compared with other existing classification methods.     

Object Bank: An Object-Level Image Representation for High-Level Visual Recognition

It is a remarkable fact that images are related to objects constituting them. In this paper, we propose to represent images by using objects appearing in them. We introduce the novel concept of object bank (OB), a high-level image representation encoding object appearance and spatial location information in images. OB represents an image based on its response to a large number of pre-trained object detectors, or ‘object filters’, blind to the testing dataset and visual recognition task. Our OB representation demonstrates promising potential in high level image recognition tasks. It significantly outperforms traditional low level image representations in image classification on various benchmark image datasets by using simple, off-the-shelf classification algorithms such as linear SVM and logistic regression. In this paper, we analyze OB in detail, explaining our design choice of OB for achieving its best potential on different types of datasets. We demonstrate that object bank is a high level representation, from which we can easily discover semantic information of unknown images. We provide guidelines for effectively applying OB to high level image recognition tasks where it could be easily compressed for efficient computation in practice and is very robust to various classifiers.     

Text Retrieval from Document Images Based on Word Shape Analysis

In this paper, we propose a method of text retrieval from document images using a similarity measure based on word shape analysis. We directly extract image features instead of using optical character recognition. Document images are segmented into word units and then features called vertical bar patterns are extracted from these word units through local extrema points detection. All vertical bar patterns are used to build document vectors. Lastly, we obtain the pair-wise similarity of document images by means of the scalar product of the document vectors. Four corpora of news articles were used to test the validity of our method. During the test, the similarity of document images using this method was compared with the result of ASCII version of those documents based on the N-gram algorithm for text documents.     

Cell morphology classification and clutter mitigation in phase-contrast microscopy images using machine learning

We propose using machine learning techniques to analyze the shape of living cells in phase-contrast microscopy images. Large scale studies of cell shape are needed to understand the response of cells to their environment. Manual analysis of thousands of microscopy images, however, is time-consuming and error-prone and necessitates automated tools. We show how a combination of shape-based and appearance-based features of fibroblast cells can be used to classify their morphological state, using the Adaboost algorithm. The classification accuracy of our method approaches the agreement between two expert observers. We also address the important issue of clutter mitigation by developing a machine learning approach to distinguish between clutter and cells in time-lapse microscopy image sequences.     

基于多角度旋转积分图的手背静脉身份识别

随着生物特征识别技术水平的飞速发展,手背静脉识别也广泛运用于各个领域。由于采集终端硬件设备和采集环境的差异,会降低识别的准确性。针对手背静脉图像在亮度,旋转,尺寸等方面造成的影响,,提出了基于多角度旋转积分图的和离散余弦变换的手背静脉图像识别方法,并进行参数优化。本文结合梯度增强的静脉图像分割方法,选取最佳角度间隔做旋转积分运算,然后通过二维离散余弦变换(DCT)截取最佳特征矩阵用做分类识别,识别率超过99.9％。 实验通过对比其它传统算法对手背静脉图像的识别效果来验证本文特征提取方法的可行性和优越性。     

Tissue classification based on 3D local intensity structures forvolume rendering

This paper describes a novel approach to tissue classification using three-dimensional (3D) derivative features in the volume rendering pipeline. In conventional tissue classification for a scalar volume, tissues of interest are characterized by an opacity transfer function defined as a one-dimensional (1D) function of the original volume intensity. To overcome the limitations inherent in conventional 1D opacity functions, we propose a tissue classification method that employs a multidimensional opacity function, which is a function of the 3D derivative features calculated from a scalar volume as well as the volume intensity. Tissues of interest are characterized by explicitly defined classification rules based on 3D filter responses highlighting local structures, such as edge, sheet, line, and blob, which typically correspond to tissue boundaries, cortices, vessels, and nodules, respectively, in medical volume data. The 3D local structure filters are formulated using the gradient vector and Hessian matrix of the volume intensity function combined with isotropic Gaussian blurring. These filter responses and the original intensity define a multidimensional feature space in which multichannel tissue classification strategies are designed. The usefulness of the proposed method is demonstrated by comparisons with conventional single-channel classification using both synthesized data and clinical data acquired with CT (computed tomography) and MRI (magnetic resonance imaging) scanners. The improvement in image quality obtained using multichannel classification is confirmed by evaluating the contrast and contrast-to-noise ratio in the resultant volume-rendered images with variable opacity values     

Segmentation-based multi-class semantic object detection

In this paper we study the problem of the detection of semantic objects from known categories in images. Unlike existing techniques which operate at the pixel or at a patch level for recognition, we propose to rely on the categorization of image segments. Recent work has highlighted that image segments provide a sound support for visual object class recognition. In this work, we use image segments as primitives to extract robust features and train detection models for a predefined set of categories. Several segmentation algorithms are benchmarked and their performances for segment recognition are compared. We then propose two methods for enhancing the segments classification, one based on the fusion of the classification results obtained with the different segmentations, the other one based on the optimization of the global labelling by correcting local ambiguities between neighbor segments. We use as a benchmark the Microsoft MSRC-21 image database and show that our method competes with the current state-of-the-art.     

基于支持向量机的纸张缺陷图像分类识别

根据支持向量机（SVM）在小样本、高维模式分类中具有的优良分类性能,提出将支持向量机应用于实际的纸张缺陷分类。针对三种现场易出现的缺陷,通过对缺陷图像进行预处理、特征选择,再利用SVM进行分类,利用交叉验证进行参数和模型选取,取得了较好的分类效果,为纸张缺陷的分类指出一种可行的方法。     

Improved segmentation of semiconductor defects using area sieves

This paper aims at investigating a novel non-referential solution to the problem of defect detection on semiconductor wafer-die images. The suggested solution focuses on segmenting defects from the images using wavelet transformation and morphology-related properties of the associated wavelet coefficients. More specifically, a novel methodology is investigated for segmenting defects by applying an area sieves technique to innovative multidimensional wavelet-based features. These features are extracted from the original defective image using the non-reference K-Level 2-D DWT (Discrete Wavelet Transform). The results of the proposed methodology are illustrated in defective die images where the defective areas are segmented with higher accuracy than the one obtained by applying other reference-based feature extraction methodologies. The first uses all the wavelet coefficients derived from the K-Level 2-D DWT, while the second one uses area sieves to segment the defective regions. Both methods involve in the same classification stage as the proposed feature extraction approach. The promising results obtained outline the importance of judicious selection and processing of 2-D DWT wavelet coefficients for industrial pattern recognition applications.     

FMI image based rock structure classification using classifier combination

Formation Micro Imager (FMI) can directly reflect changes of wall stratums and rock structures, and is an important factor to classify stratums and identify lithology for the oil and gas exploration. Conventionally, people analyze FMI images mainly with manual processing, which is, however, extremely inefficient and incurs a heavy workload for experts. In this paper, we propose an automatic rock structure classification system using image processing and pattern recognition technologies. We investigate the characteristics of rock structures in FMI images carefully. We also develop an effective classification framework with classifier combination that can integrate the domain knowledge from experienced geologists successfully. Our classification system includes three main steps. First, various effective features, specially designed for FMI images, are calculated and selected. Then, the corresponding single classifier associated with each feature is constructed. Finally, all these classifiers are combined as an effective cascade recognition system. We test our rock structure classification system with real FMI rock images. In experiments, with only one training sample per class, the average recognition accuracy of our proposed system is 81.11%. The accuracy is 15.55 percent higher than the traditional 1-nearest neighborhood method. Moreover, this automatic system can significantly reduce the complexity and difficulty in the rock structure analysis task for the oil and gas exploration.     

Hierarchical fusion of multi-spectral face images for improved recognition performance

This paper presents a two level hierarchical fusion of face images captured under visible and infrared light spectrum to improve the performance of face recognition. At image level fusion, two face images from different spectrums are fused using DWT based fusion algorithm. At feature level fusion, the amplitude and phase features are extracted from the fused image using 2D log polar Gabor wavelet. An adaptive SVM learning algorithm intelligently selects either the amplitude or phase features to generate a fused feature set for improved face recognition. The recognition performance is observed under the worst case scenario of using single training images. Experimental results on Equinox face database show that the combination of visible light and short-wave IR spectrum face images yielded the best recognition performance with an equal error rate of 2.86%. The proposed image-feature fusion algorithm also performed better than existing fusion algorithms.     

A spatial-channel progressive fusion ResNet for remote sensing classification

In recent years, the panchromatic (PAN) and the multispectral (MS) remote sensing images classification has become a research hotspot. In this paper, we propose a spatial-channel progressive fusion residual network (SCPF-ResNet) for multi-resolution remote sensing classification. Firstly, for the inputs of the proposed network, the interactive data fusion strategy (IDFS) combines generalized-intensity-hue-saturation (GIHS), and discrete wavelet transform (DWT) to interfuse patch pairs of the PAN and the MS images, so as to increase the similarity between them, thus reduce the difference in information between them. Secondly, for the branches of feature extraction, we design an adaptive spatial attention module (ASA-Module) and an adaptive channel attention module (ACA-Module) to strengthen spatial features from both larger-sized with smaller-sized targets and spectral features among channels. Finally, we insert the ASA-Module and ACA-Module into the residual modules to form a triple-branch network and use the common spatial-channel features extracted by the Fusion_Branch to gradually enhance the pure independent features extracted by the PAN_Branch and the MS_Branch, respectively. The experimental results indicate that SCPF-ResNet can achieve competitive performance.     

基于傅里叶变换特征遗传算法的人脸识别

提出了一种基于傅里叶变换特征遗传算法的人脸识别方法。首先对人脸图像进行傅里叶变换,并去掉不包含识别信息的零频率分量。然后以人脸图像傅里叶变换的模作为描述人脸模式的特征,考虑到并不是所有傅里叶特征都包含较好的分类信息,应用遗传算法进行特征选取。实验结果表明：给出的方法不仅具有较好的识别性能,而且可以在一定程度上减少人脸模式的存储空间。     

Multi-stage classification of emotional speech motivated by a dimensional emotion model

This paper deals with speech emotion analysis within the context of increasing awareness of the wide application potential of affective computing. Unlike most works in the literature which mainly rely on classical frequency and energy based features along with a single global classifier for emotion recognition, we propose in this paper some new harmonic and Zipf based features for better speech emotion characterization in the valence dimension and a multi-stage classification scheme driven by a dimensional emotion model for better emotional class discrimination. Experimented on the Berlin dataset with 68 features and six emotion states, our approach shows its effectiveness, displaying a 68.60% classification rate and reaching a 71.52% classification rate when a gender classification is first applied. Using the DES dataset with five emotion states, our approach achieves an 81% recognition rate when the best performance in the literature to our knowledge is 76.15% on the same dataset.