利用对抗网络改进多标记图像分类

为了更有效地对多标记图像进行分类,提出一个改进的卷积神经网络模型,通过融合多层次特征并利用空间金字塔池化来学习多标记图像中的多尺度特征,同时设计对抗网络生成新的样本辅助模型训练.首先,对传统卷积神经网络模型进行改进,利用空间金字塔池化层替换网络的最后一层,并将在ImageNet上预先训练好的参数传递给该模型;然后,通过将深层特征和浅层特征进行融合,使得模型对不同尺度的物体具有更好的识别能力;最后,设计了一个对抗网络生成带遮挡的样本,使模型对遮挡物体的识别也具有良好的鲁棒性.实验测试在2个基准数据集上进行,文中模型在Corel5K数据集上的平均查准率和平均查全率分别为0.457和0.427,mAP值达到0.442,而在PASCAL VOC 2012数据集上的mAP值则达到0.85.实验结果表明,与当前国际先进的模型相比,该模型具有更好的有效性和更强的鲁棒性.     

A robust technique based on invariant moments – ANFIS for recognition of human parasite eggs in microscopic images

In this study, we propose a robust technique based on invariant moments – adaptive network based fuzzy inference system (IM-ANFIS). In this technique, some digital image processing methods such as noise reduction, contrast enhancement, segmentation, and morphological process are used for feature extraction stage of IM-ANFIS approach used in this study. Recently, the pattern recognition principles have come into prominence. The pattern recognition includes operation and design of systems that recognize patterns in data sets. Important application areas of pattern recognition techniques are character recognition, speech analysis, image segmentation, man and machine diagnostics and industrial inspection. The technique presented in this study enables to classify 16 different parasite eggs from their microscopic images. This proposed recognition method includes three stages. In first stage, a preprocessing subsystem is realized for obtaining unique features from the same group of patterns. In second stage, a feature extraction mechanism which is based on the invariant moments is used. In third stage, an adaptive network based fuzzy inference system (ANFIS) classifier is used for recognition process. We conduct computer simulations on MATLAB environment. The overall success rate is almost 95%.     

Rotation-invariant neural pattern recognition system withapplication to coin recognition

In pattern recognition, it is often necessary to deal with problems to classify a transformed pattern. A neural pattern recognition system which is insensitive to rotation of input pattern by various degrees is proposed. The system consists of a fixed invariance network with many slabs and a trainable multilayered network. The system was used in a rotation-invariant coin recognition problem to distinguish between a 500 yen coin and a 500 won coin. The results show that the approach works well for variable rotation pattern recognition.     

A multilayered self-organizing artificial neural network forinvariant pattern recognition

An artificial neural network that self-organizes to recognize various images presented as a training set is described. One application of the network uses multiple functionally disjoint stages to provide pattern recognition that is invariant to translations of the object in the image plane. The general form of the network uses three stages that perform the functionally disjoint tasks of preprocessing, invariance, and recognition. The preprocessing stage is a single layer of processing elements that performs dynamic thresholding and intensity scaling. The invariance stage is a multilayered connectionist implementation of a modified Walsh-Hadamard transform used for generating an invariant representation of the image. The recognition stage is a multilayered self-organizing neural network that learns to recognize the representation of the input image generated by the invariance stage. The network can successfully self-organize to recognize objects without regard to the location of the object in the image field and has some resistance to noise and distortions     

A topological and temporal correlator network for spatiotemporalpattern learning, recognition, and recall

In this paper, we describe the design of an artificial neural network for spatiotemporal pattern recognition and recall. This network has a five-layered architecture and operates in two modes: pattern learning and recognition mode, and pattern recall mode. In pattern learning and recognition mode, the network extracts a set of topologically and temporally correlated features from each spatiotemporal input pattern based on a variation of Kohonen's self-organizing maps. These features are then used to classify the input into categories based on the fuzzy ART network. In the pattern recall mode, the network can reconstruct any of the learned categories when the appropriate category node is excited or probed. The network performance was evaluated via computer simulations of time-varying, two-dimensional and three-dimensional data. The results show that the network is capable of both recognition and recall of spatiotemporal data in an online and self-organized fashion. The network can also classify repeated events in the spatiotemporal input and is robust to noise in the input such as distortions in the spatial and temporal content.     

Improving optical Fourier pattern recognition by accommodating the missing information

Information is vital to pattern recognition, yet we seldom have enough of it. In fact, some ignorance (absence of knowledge) is inevitable when we try to learn how to classify any real objects or events by learning from a finite set of exemplars. Overcoming that ignorance requires special strategies that are outlined here. The net result is that optical Fourier pattern recognition is converted from a very weak discriminator to the most powerful of all in terms of its generalization ability. The analysis is done on a very simple problem, so the logic can be understood visually as well as mathematically.     

Model-based detection,segmentation, and classification for image analysis using on-line shape learning

Detection, segmentation, and classification of specific objects are the key building blocks of a computer vision system for image analysis. This paper presents a unified model-based approach to these three tasks. It is based on using unsupervised learning to find a set of templates specific to the objects being outlined by the user. The templates are formed by averaging the shapes that belong to a particular cluster, and are used to guide a probabilistic search through the space of possible objects. The main difference from previously reported methods is the use of on-line learning, ideal for highly repetitive tasks. This results in faster and more accurate object detection, as system performance improves with continued use. Further, the information gained through clustering and user feedback is used to classify the objects for problems in which shape is relevant to the classification. The effectiveness of the resulting system is demonstrated in two applications: a medical diagnosis task using cytological images, and a vehicle recognition task. Received: 5 November 2000 / Accepted: 29 June 2001 Correspondence to: K.-M. Lee     

基于边缘学习的低分辨率图像识别算法

由于受光照条件、拍摄角度、传输设备以及周围环境的影响,刑侦视频图像中的目标物体往往分辨率较低,难以识别。针对低分辨率图像识别问题,在经典LeNet-5识别网络的基础上,提出了一种基于边缘学习的低分辨率图像识别算法。首先由边缘生成对抗网络生成低分辨率图像的幻想边缘,该边缘与高分辨率图像边缘相近;再将该低分辨图像的生成边缘信息作为先验信息融合到识别网络中对低分辨率图像进行识别。在MNIST、EMNIST和Fashion-mnist三个数据集上分别进行实验,结果表明,将低分辨图像的幻想边缘信息融合到识别网络中可以提高低分辨率图像的识别率。     

高分辨卫星图像卷积神经网络分类模型

目的 卫星图像往往目标、背景复杂而且带有噪声,因此使用人工选取的特征进行卫星图像的分类就变得十分困难。提出一种新的使用卷积神经网络进行卫星图像分类的方案。使用卷积神经网络可以提取卫星图像的高层特征,进而提高卫星图像分类的识别率。方法 首先,提出一个包含六类图像的新的卫星图像数据集来解决卷积神经网络的有标签训练样本不足的问题。其次,使用了一种直接训练卷积神经网络模型和3种预训练卷积神经网络模型来进行卫星图像分类。直接训练模型直接在文章提出的数据集上进行训练,预训练模型先在ILSVRC（the ImageNet large scale visual recognition challenge）-2012数据集上进行预训练,然后在提出的卫星图像数据集上进行微调训练。完成微调的模型用于卫星图像分类。结果 提出的微调预训练卷积神经网络深层模型具有最高的分类正确率。在提出的数据集上,深层卷积神经网络模型达到了99.50%的识别率。在数据集UC Merced Land Use上,深层卷积神经网络模型达到了96.44%的识别率。结论 本文提出的数据集具有一般性和代表性,使用的深层卷积神经网络模型具有很强的特征提取能力和分类能力,且是一种端到端的分类模型,不需要堆叠其他模型或分类器。在高分辨卫星图像的分类上,本文模型和对比模型相比取得了更有说服力的结果。     

基于组合不变矩和神经网络的三维物体识别

在三维物体识别系统中,提出将三维物体的Hu不变矩和仿射不变矩两者的低阶矩组合作为三维物体的特征,结合改进的BP神经网络应用于三维物体的分类识别。理论分析和仿真实验表明组合这两种矩特征进行物体识别,性能优于单独使用Hu不变矩,如果进一步对这两种组合的矩特征进行主成分分析处理,可显著提高系统识别性能,并减少网络的训练时间。     

Multi-scale crown closure retrieval for moso bamboo forest using multi-source remotely sensed imagery based on geometric-optical and Erf-BP neural network models

This article focuses on retrieving the multi-scale crown closure (CC) of Moso bamboo forest using Système Pour l’Observation de la Terre (SPOT5) and Landsat Thematic Mapper (TM) satellite remotely sensed imagery based on the geometric-optical model and the artificial neural network (ANN) model. CC at local scale was first retrieved using the Li-Strahler geometric-optical model (LSGM) and images from an unmanned aerial vehicle (UAV). Then, multi-scale CC was retrieved using the Erf-BP model (a kind of back-propagation (BP) feed-forward neural network, which takes a Gaussian error function (Erf) as an activation function of the hidden layer) based on a combination of SPOT5 and Landsat TM images. The results show that by combining multi-source remotely sensed data, the CC of Moso bamboo forest can be retrieved at the local region, township area, and county scale with high accuracy using the Erf-BP model. Estimated values have a linear relationship with the observed values at a significance level of 0.05. The highest accuracy of the retrieval of CC (referred to as LSGM-UAV-CC) was observed at the local region based on LSGM and UAV, with the coefficient of determination (R²) of 0.63, followed by that at the township area with an R² of 0.0.55 based on LSGM-UAV-CC and SPOT5 data using the Erf-BP model (Erf-BP-SPOT5-CC), and that at the county scale with an R² of 0.54 based on Erf-BP-SPOT5-CC and Landsat TM data using the Erf-BP model (Erf-BP-TM-CC).     

Dimension Reduction and Construction of Feature Space for Image Pattern Recognition

We mathematically and experimentally evaluate the validity of dimension-reduction methods for the computation of similarity in image pattern recognition. Image pattern recognition identifies instances of particular objects and distinguishes differences among images. This recognition uses pattern recognition techniques for the classification and categorisation of images. In numerical image pattern recognition techniques, images are sampled using an array of pixels. This sampling procedure derives vectors in a higher-dimensional metric space from image patterns. To ensure the accuracy of pattern recognition techniques, the dimension reduction of the vectors is an essential methodology since the time and space complexities of processing depend on the dimension of the data. Dimension reduction causes information loss of topological and geometrical features of image patterns. Through both theoretical and experimental comparisons, we clarify that dimension-reduction methodologies that preserve the topology and geometry in the image pattern space are essential for linear pattern recognition. For the practical application of methods of dimension reduction, the random projection works well compared with downsampling, the pyramid transform, the two-dimensional random projection, the two-dimensional discrete cosine transform and nonlinear multidimensional scaling if we have no a priori information on the input data.     

基于SVM的车型检测和识别算法

根据模式识别理论和支持向量机(SVM)网络技术,对运动车辆的检测和模式识别、分类进行了研究,提出了基于双帧差“或”运算检测法和基于SVM网络的车型识别和分类算法。实验结果表明,所设计的检测方法和SVM模式识别方法能够快速有效地识别车辆类型并正确地进行分类。     

基于改进的卷积神经网络光学薄膜缺陷识别

为对光学薄膜缺陷图像进行准确识别分类,提出一种基于改进的卷积神经网络光学薄膜缺陷图像识别方法。为突出输入图像中的缺陷信息,采用改进的LBP算法对图像进行预处理。从三个方面对传统的卷积神经网络进行改进:为了解决单通道卷积神经网络对图像特征提取不充分的问题,构建双通道卷积神经网络;改进传统的ReLU激活函数,避免模型出现欠拟合现象;使用支持向量机(SVM)代替Softmax分类器,提高计算效率和准确率。光学薄膜缺陷图像仿真识别实验表明,所提方法分类平均准确率高达93.2%,训练时间为964 s,充分验证了所提方法的鲁棒性和有效性。     

Global feature space neural network for active computer vision

We advance new active computer vision algorithms based on the Feature space Trajectory (FST) representations of objects and a neural network processor for computation of distances in global feature space. Our algorithms classify rigid objects and estimate their pose from intensity images. They also indicate how to automatically reposition the sensor if the class or pose of an object is ambiguous from a given viewpoint and they incorporate data from multiple object views in the final object classification. An FST in a global eigenfeature space is used to represent 3D distorted views of an object. Assuming that an observed feature vector consists of Gaussian noise added to a point on the FST, we derive a probability density function for the observation conditioned on the class and pose of the object. Bayesian estimation and hypothesis testing theory are then used to derive approximations to the maximum a posterioriprobability pose estimate and the minimum probability of error classifier. Confidence measures for the class and pose estimates, derived using Bayes theory, determine when additional observations are required, as well as where the sensor should be positioned to provide the most useful information.     

纹理图象亮度阈值法提取SAR图象居民地

由于微波辐射的复杂特性，从合成孔径雷达图象上提取类似于居民地复杂结构的目标物的研究仍处于探索中．通过研究居民地对合成孔径雷达(SAR)的微波散射特性，分析居民地在SAR图象上的纹理特征，综合利用纹理分析、模式识别和颜色空间变换技术，提出了一种新的提取雷达图象上居民地的方法．该方法在共生矩阵纹理分析的基础上，选取3个合适的特征分量合成彩色纹理特征图象，再通过HIS变换获得亮度分量，使用亮度阈值分割图象来提取出居民地．此方法的特点是，其受雷达系统影响较小，适应性较强，以二值图象的形式记录居民地的提取结果．试验表明，利用此方法在SAR图象上提取居民地具有70％以上的正确识别率．