长度归一化扫描的合成孔径雷达图像船舶尾迹恒虚警率检测

提出一种基于长度归一化扫描的合成孔径雷达(SAR)图像船舶尾迹检测算法.存在距离向运动分量的船舶在SAR图像上会发生方位向偏移,那么尾迹的起点必在方位向上这一偏移量范围内.根据这一物理事实,将尾迹检测的搜索范围限定在可能为尾迹的线段上,从而提高了检测效率.算法通过利用线性积分和长度归一化这两个方法将矩形滑动窗口下的线性特征检测转化为了点特征检测,并通过经典的虚警率(CFAR)检测理论实现检测结果的输出.利用COSMO-SkyMed数据对该算法进行了实验验证,实验结果表明,该算法在尾迹检测上具有检测能力强、速度快的优点,船舶速度反演具有较高精度.     

基于Laplacian Eigenmap的图像变化检测虚警优化技术

对点目标的图像变化检测,现有的变化检测技术结果往往存在着虚警过大的问题。通过深入分析多个传统的变化检测方法的特点,利用各方法的互补性,提出了利用Laplacian Eigenmap对多个方法检测结果进行降维分类的优化技术。首先把各个方法对某个像素的检测结果用向量的形式进行表示,然后利用Laplacian Eigenmap对整个图像的数据流形在低维空间展开,最后利用模糊分类进行分类。该技术有两个优势：（1）在保证现有较高检测率的同时,大大降低了结果的虚警率;（2）它极大地降低了在传统方法中由于人为阈值取舍带来的偏差风险。但该技术的不足之处是增加了计算量。     

一种改进的深度残差网络行人检测方法

为了提高行人检测方法的准确率,针对行人图像特征,提出一种基于深度残差网络和YOLO(You Only Look Once)方法的行人检测方法。以加强行人特征表达为目的,通过分析行人在图像中的表达和分布特征,提出一种不影响实时性的矩形输入深度残差网络分类模型以改进YOLO检测方法,使模型能够更好的表征行人;为了进一步提高模型的准确率和泛化能力,采用了混合行人数据集训练的方式,提取VOC数据集的行人数据与INRIA数据集组成混合数据集进行训练,明显降低了漏检率;并且利用聚类分析预测框的方法重新设计了初始预测框,提高行人定位能力并加快收敛。经公开的INRIA数据集的测试实验证明,本方法较主流的行人检测方法每张图片误检率有明显改善,降低至13.86%,有1.51%至58.62%不同程度的提升,并且本方法拥有良好的实时性和泛化能力,实用性强。     

Constant false alarm rate detection in Pareto Type II clutter

Recent investigations have shown that the Pareto class of models provide a valid approximation for the statistical structure of the backscattering from the sea, for high resolution X-band maritime surveillance radar. This has stimulated the research and development of non-coherent radar detection processes for operation in such a clutter model environment. Using data from Defence Science and Technology Group's X-band radar the application of a Pareto Type I clutter model has been justified, which has facilitated the development of sliding window decision processes. However it has been found that when these detectors are applied to synthetic target detection in real data there are some issues resulting in substantial detection losses. In order to rectify this it is necessary to investigate the development of radar detection schemes under a Pareto Type II clutter model assumption. Using a transformation approach for radar detector design it will be shown that it is possible to construct detection processes that achieve the constant false alarm rate property with respect to the Pareto shape parameter, as in the Pareto Type I case, while requiring a priori knowledge of the Pareto scale parameter. Performance analysis includes application to real X-band clutter returns with synthetic target models.     

机场跑道异物监测雷达的杂波图恒虚警率检测

随天线扫描平稳变化的强地杂波是实现机场跑道异物(FOD)检测的主要干扰,传统的空域恒虚警率(CFAR)处理不能有效地检测到目标,针对上述问题,提出了一种单元平均杂波图恒虚警率检测算法。首先基于系统特性和跑道环境建立了回波信号模型;然后通过杂波图划分、单元平均、递归滤波等处理技术,实现了距离—方位二维恒虚警率检测;最后进一步分析影响检测性能的主要参数。仿真结果表明,所提算法在低信杂比背景下能有效检测到弱目标,并获得较高的检测概率。     

Constant false alarm rate detector based on the maximal reference cell

In order to improve the detection performance of constant false alarm rate (CFAR) detectors in multiple targets situations, a CFAR detector based on the maximal reference cell (MRC) named MRC-CFAR is proposed. In MRC-CFAR, a comparison threshold is generated by multiplying the amplitude of MRC by a scaling factor. The number of the reference cells left, whose amplitudes are smaller than the comparison threshold, is counted and compared with a threshold integer. Based on the comparison result, proper reference cells are selected for detection threshold computation. A closed-form analysis for MRC-CFAR in both homogeneous and non-homogeneous environments is presented. The performance of MRC-CFAR is evaluated and compared with other CFAR detectors. MRC-CFAR exhibits a very low CFAR loss in a homogeneous environment and performs robustly during clutter power transitions. In multiple targets situations, MRC-CFAR achieves a much better detection performance than switching CFAR (S-CFAR) and order-statistic CFAR (OS-CFAR). Experiment results from an X-band linear frequency modulated continuous wave radar system are given to demonstrate the efficiency of MRC-CFAR. Because ranking reference cells is not required for MRC-CFAR, the computation load of MRC-CFAR is low; it is easy to implement the detector in radar system in practice.     

The constant false alarm rate property in transformed noncoherent detection processes

Recently a transformation approach for noncoherent radar detector design has been introduced, where the classical constant false alarm rate detectors for Exponentially distributed clutter are modified to operate in any clutter intensity model of interest. Recent applications of this approach have introduced new decision rules for target detection in Pareto and Weibull distributed clutter. These transformed detectors tended to lose the constant false alarm rate property with respect to one of the clutter parameters. A closer examination of this transformation process yields conditions under which the constant false alarm rate property can be retained. Based upon this, a new model for X-band maritime radar returns is investigated, and corresponding detectors are developed. The relative merits of this new development are investigated with synthetic and real X-band data.     

Constant false alarm rate detection of saccadic eye movements in electro-oculography

The analysis of eye movements has proven to be valuable in both clinical work and research as well as in other fields besides medicine. The detection of saccadic eye movements and the extraction of related saccade parameters, such as maximum angular velocity, amplitude, and duration, are usually performed during the analysis of electro-oculographic (EOG) signals. This article considers a saccade detection method that is based on the constant false alarm rate technique, in which the detection sensitivity is continuously adjusted on the basis of the observed signal in order to keep the number of false alarms constant. The method is computationally efficient, it can operate autonomously without user intervention, and it is capable of detecting saccades in a sequential fashion. Therefore, the method finds potential use in applications that require automated analysis of electro-oculographic signals. Because of the constant false alarm rate property, the method can also perform in situations where ideal measurement conditions cannot be guaranteed and noise presents a considerable problem.     

A novel lossless compression framework for facial depth images in expression recognition

With the development of AR and VR, depth images are widely used for facial expression analysis and recognition. To reduce the storage size and save bandwidth, an efficient compression framework is desired. In this paper, we propose a novel lossless compression framework for facial depth images in expression recognition. In the proposed framework, two steps are designed to remove the redundancy in the facial depth images, which are data preparing and bitstream encoding operations. In the data preparing operation, the original image is represented by the same and different parts between the left and right sides. In the bitstream encoding operation, these parts are compressed to get the final bitstream. The proposed framework is implemented and examined on the BU-3DFE Database. Experimental result shows that the proposed technique outperforms existing lossless compression frameworks in terms of compression efficiency, and the average data size is reduced to 25.27% by the proposed framework.

     

A novel deep learning framework for copy-moveforgery detection in images

Multimedia Tools and Applications - In this era of technology, digital images turn out to be ubiquitous in a contemporary society and they can be generated and manipulated by a wide variety of...     

基于双目图像的行人检测与定位系统研究

针对半全局匹配算法（Semi-Global Matching,SGM）的视差图匹配度较低的问题,以及目标检测算法Fast-YOLO对小目标的检测能力不足的问题,提出一种基于双目图像的行人检测与定位系统。系统首先利用图像分割块具有视差相似性的特点,使用基于快速图像分割的SGM算法对双目图像进行立体匹配,然后修改Fast-YOLO网络模型,提高网络分辨率,使用改进的Fast-YOLO网络进行行人检测。实验结果表明,基于快速图像分割的SGM算法较好地解决了匹配度较低问题,基于Fast-YOLO改进的行人检测网络明显地提高了对小目标的检测能力。系统实现了对行人的检测和定位,并使用GPU达到实时的计算效率。     

A versatile hardware architecture for a constant false alarm rate processor based on a linear insertion sorter

Constant False Alarm Rate (CFAR) algorithms are used in digital signal processing applications to extract targets from background in noisy environments. Some examples of applications are target detection in radar environments, image processing, medical engineering, power quality analysis, features detection in satellite images, Pseudo-Noise (PN) code detectors, among others. This paper presents a versatile hardware architecture that implements six variants of the CFAR algorithm based on linear and nonlinear operations for radar applications. Since some implemented CFAR algorithms require sorting the input samples, a linear sorter based on a First In First Out (FIFO) schema is used. The proposed architecture, known as CFAR processor, can be used as a specialized module or co-processor for Software Defined Radar (SDR) applications. The results of implementing the CFAR processor on a Field Programmable Gate Array (FPGA) are presented and discussed.     

A novel multiplex cascade classifier for pedestrian detection

Reliable pedestrian detection is of great importance in visual surveillance. In this paper, we propose a novel multiplex classifier model, which is composed of two multiplex cascades parts: Haar-like cascade classifier and shapelet cascade classifier. The Haar-like cascade classifier filters out most of irrelevant image background, while the shapelet cascade classifier detects intensively head-shoulder features. The weighted linear regression model is introduced to train its weak classifiers. We also introduce a structure table to label the foreground pixels by means of background differences. The experimental results illustrate that our classifier model provides satisfying detection accuracy. In particular, our detection approach can also perform well for low resolution and relatively complicated backgrounds.     

Accurate saliency detection based on depth feature of 3D images

This paper presents an accurate saliency detection algorithm customized for 3D images which contain abundant depth cue. Firstly, depth feature is calculated based on the sharp regions’ positions within the focal stack. Then, we compute the coarse saliency map by subtracting the background region from the all-focus image according to the depth feature. Finally, we employ the contrast information in the coarse saliency map to obtain the final result. Experiments on light field dataset demonstrate that our approach favorably outperforms five state-of-the-art methods in terms of precision, recall and F-Measure. Moreover, the depth feature is validated to be a valuable complement to existing visual saliency analysis under the circumstance that the background regions are complex or similar to salient object regions.     

基于快速级联分类器的行人检测方法研究

行人检测是图像处理、计算机视觉等方面研究的重要环节,通常用于视频监控和智能车辆等领域。行人检测图像易受到背景的影响,常用的帧差法及单纯训练分类器法在行人检测中存在着准确率低、分类训练算法复杂、实时性差等问题。首先采用改进型帧差法获取行人运动信息,然后利用直方图坐标对应划分出运动区域,最后通过训练双特征级联分类器对运动区域进行检测识别。实验结果表明,本方法可以有效减少误检和漏检现象,检测时间平均减少了32.77ms,检测准确率平均提高了10%以上,因此本方法有效提高了识别准确率和识别速度。     

Tracking a hand in interaction with an object based on single depth images

Multimedia Tools and Applications - Tracking a hand in interaction with an object based on vision is a challenging research topic. The occlusions that occur during the hand-object interaction make...     

A novel approach based on structural information for change detection in SAR images

Structural information, extracted by simulating the human visual system (HVS), is independent of viewing conditions and individual observers. Structural similarity (SSIM), a measure of similarity between two images, has been widely used in image quality assessment. Given the fact that the change detection techniques identify the changed area by the similarity of multi-temporal images, SSIM has significant prospect in change detection of synthetic aperture radar (SAR) images. However, the experimental results show that SSIM performs worse in change detection of multi-temporal SAR images. In this study, we first propose an advanced SSIM (ASSIM) based on a two-step assumption of extracting structural information and a visual attention measure (VAM) model. Then, we propose a novel approach based on ASSIM for change detection in SAR images. SSIM, ASSIM, and state-of-the-art methods are tested on two datasets to compare their performances in change detection of SAR images. Experimental results show that the proposed method can acquire a better difference image than SSIM and other state-of-the-art methods, and improve the accuracy of change detection in SAR images effectively.     

A new approach for oil tank detection using deep learning features with control false alarm rate in high-resolution satellite imagery

ABSTRACT

Oil tank detection is a challenging task, primarily due to high time-consumption. This paper aims at further investigating this challenge and proposes a new hierarchical approach to detect oil tanks, especially with respect to how false alarm rates are reduced. The proposed approach is divided into four stages: region of interest (ROI) extraction, circular object detection, feature extraction, and classification. The first stage, which is a key component of this approach to reduce false alarm and processing time, is applied by an improved faster region-based convolutional neural network (Faster R-CNN) to extract oil depots. In the second stage, a number of candidate objects of the target are selected from the extracted ROIs by a fast circle detection method. Afterwards, in the third stage, a robust feature extractor based on a combination of the output feature vectors from convolutional neural network (CNN), as a high-level feature extractor, and histogram of oriented gradients (HOG), as a low-level feature extractor, are used for representing features of various targets. Finally, the support vector machine (SVM) is employed for classification. The experimental results confirm that the proposed approach has good prediction accuracy and is able to reduce the false alarm rates.     

A two-stage anomaly detection framework: Towards low omission rate in industrial vision applications

In industrial manufacturing, there are many types of defective samples that are difficult to obtain. Practical industrial vision anomaly detection has proven to be a challenging task because techniques use only normal (non-defective) samples to train a model to detect anomalies. Currently, some reasonably effective models do not perform very well once differences between samples are large, and they ignore the fact that the cost of missing a defect is much higher than the cost of misidentifying a normal sample. To that end, in this paper, we propose a two-stage framework to construct an anomaly detector. We first train a classification network and then build a one-class classifier on learned representations using another pre-trained network. This paper innovatively proposes using the theoretical quantile as the discriminant threshold. We conduct experiments on the Nut and Motor Brush Holder datasets from real industrial production lines. The results show that our method greatly reduces missed detection of anomalous samples, achieving state-of-the-art AUROC scores of 99.3 % and 96.2 %. We also conduct experiments on the publicly available dataset Rd-MVTec AD, showing that our model has good generalizability and fast testing speed while maintaining high AUROC scores. Our model gives excellent results for nonaligned and defective data with diverse anomalous patterns, and it is easy to optimize. Therefore, not only does our technique handle industrial cold starts well, but it also meets the requirement of online updating, which indicates that our solution is highly suitable for industrial manufacturing scenarios.     

基于半监督学习的行人检测方法研究

本文提出了基于半监督学习的行人检测方法,用以解决大量的无标记样本问题。在集成分类器的训练过程中,选择BP神经网络分类器、SVM分类器和KNN分类器作为3个子分类器,利用协同训练机制对各个子分类器进行协同训练。针对半监督学习中误标记样本问题,引入富信息策略和辅助学习策略消除训练过程引入的噪声,同时充分利用无标记样例,进而提高分类器的分类精度。通过对测试集和实时视频进行的行人检测实验,证明了本文方法的可行性和有效性。