一种串联组合分类的红外行人检测方法

针对红外图像行人检测中虚警率高的问题,提出了一种串联组合分类的红外行人检测方法。该方法利用感兴趣区域的宽高比和方向梯度直方图2种特征,采用串联组合方式实现分类识别。实验结果中检测率达到了90.5%,虚警率减少了43.7%,表明提出的方法可有效检测行人同时降低虚警。     

基于HOG纹理的全天时十字路口车尾检测算法

HOG纹理因其良好的鲁棒性,在纹理描述中被广泛使用。提出了一种将HOG纹理应用于十字路口全天候车尾检测的算法。即分别采集了白天和夜间该场景下的车尾作为正样本、非车辆和车辆的一部分作为负样本,经预处理后,提取较低维数的HOG纹理送入支持向量机进行训练,得到白天和夜间的识别模型,在检测中根据一定的条件进行切换。对多段视频进行测试证明,该种算法对不同时段的交通场景都具有较高的稳定的车尾识别率,且优于单模型的识别效果。     

基于OTSU分割和HOG特征的行人检测与跟踪方法

传统的HOG算法针对整幅图像进行行人特征提取,大量的非人窗口计算必然降低检测的准确率和效率。为此,提出一种基于OTSU分割和HOG特征的行人检测与跟踪方法。利用OTSU算法以最佳阈值分割图像,在分割区域的基础上进行Canny边缘检测,通过边缘的对称性计算确定行人候选区,继而采用经PCA方法降维后的HOG特征和隐马尔可夫模型对行人候选区进行检测验证。最后,以确定的行人区域为跟踪窗口,利用CamShift算法跟踪行人。多组实验结果证明,本文方法的行人检测效率和精度均有所提高,跟踪性能稳定、可靠。     

基于支持向量机的自动目标识别算法

基于对目标识别精确性的要求,提出了基于支持向量机的自动目标识别算法。介绍了基于支持向量机的自动目标识别系统的组成和识别流程,实现了目标的特征提取、SVM分类器的参数寻优,并将优化的SVM模型应用于未知图像的目标识别中。实验表明,该方法识别效果良好,具有较好的抗复杂背景的能力。     

空间目标自适应光学图像椭圆部件检测

为了识别空间目标的椭圆部件,提出了一种基于自适应光学图像的椭圆检测方法。首先,利用RL(RichardsonLucy)方法对自适应光学图像进行复原,在此基础上,采用弧支撑线段(Arc-Support Line Segments, ASLS)方法对复原图像进行椭圆检测。针对ASLS算法使用的Canny边缘提取算法带来的“弧段过分割”和“语义信息差”等问题,提出了基于多尺度组合分组(Multiscale Combinatorial Grouping, MCG)边缘提取的解决方法。最后,针对ASLS算法使用优度指标等验证方法存在部分虚假椭圆的情况,综合利用多种几何指标进行约束,有效地消除了虚假椭圆。实验结果表明：椭圆中心点检测误差优于3 pixels,半长轴误差优于4 pixels,方向角误差优于3°。在重叠面积门限为0.65时,本文算法的准确率为85.7%、召回率为93.3%,F值指标为0.893,优于传统椭圆检测算法。     

基于特征选择的网络入侵检测模型研究

为了有效从收集的恶意数据中选择特征去分析,保障网络系统的安全与稳定,需要进行网络入侵检测模型研究。但目前方法是采用遗传算法找出网络入侵的特征子集,再利用粒子群算法进行进一步选择,找出最优的特征子集,最后利用极限学习机对网络入侵进行分类,但该方法准确性较低。为此,提出一种基于特征选择的网络入侵检测模型研究方法。该方法首先以增强寻优性能为目标对网络入侵检测进行特征选择,结合分析出的特征选择利用特征属性的Fisher比构造出特征子集的评价函数,然后结合计算出的特征子集评价函数进行支持向量机完成对基于特征选择的网络入侵检测模型研究方法。仿真实验表明,利用支持向量机对网络入侵进行检测能有效地提高入侵检测的速度以及入侵检测的准确性。     

利用跟踪轨迹特征和SVDD-SVM联合分类器的水下慢速小目标分类方法

针对蛙人、无人水下航行器(UUV)等慢速小目标分类识别所面临的小样本、类不平衡问题,提出了利用轨迹特征、支持向量数据描述(SVDD)与支持向量机(SVM)的联合分类方法。该方法将水下慢速小目标类型简化为蛙人、UUV、其他3类,利用跟踪轨迹特征设计多维特征量,构建SVDD-SVM联合分类器获得分类结果。具体为,针对小样本问题,采用参数维度小、训练数据量要求低的SVDD、SVM作为分类器的基本单元。针对类不平衡问题,使用2个并联的单分类SVDD和1个与两者串联的二分类SVM设计联合分类器,同时为联合分类器的输出设计投票机制保证分类结果的稳健性。实测数据处理结果表明,所提SVDD-SVM联合分类器对蛙人目标的平均召回率可达86%,平均精确率可达87%;对UUV目标的平均召回率可达85%,平均精确率可达86%。所提方法在小样本、类不平衡条件下具有优于传统方法的分类准确性和稳健性。     

基于支持向量机的图像飞机目标自动识别算法研究

为了能够快速、准确地识别飞机目标,文章给出了一种基于支持向量机的飞机目标自动识别方法;采用Touzi边缘提取,得到目标形状参数的几何特征,Hu不变矩等16个特征矢量作为SVM的训练样本,通过SVM训练得到飞机目标识别模型,从而完成飞机目标的自动识别;试验结果显示,该算法对不同尺度和模糊程度的飞机目标的识别度可达99%; 该算法减少了样本训练时间,在提高识别准确率的同时降低了算法的复杂度,具有识别度高、识别速度快的特点,可用于飞机目标的快速识别。     

基于混合高斯模型的运动目标检测方法研究

为了能够及时检测到图像场景中的运动目标,提出一种基于混合高斯模型的运动目标检测方法。借助把图像的像素值看成是前景高斯分布和背景高斯分布的组合,进行了背景估计和自适应背景更新;通过对背景差分后的图像进行多目标分割,进行了多个运动目标的检测。实验发现:随着图像序列的背景不断变化,基于混合高斯模型算法能够准确估计出变化的背景,通过将场景图像和背景图像进行差分,检测到感兴趣的运动目标。     

一种基于目标与背景特征分离模型的高光谱目标检测修正算法

高光谱图像立方体数据可以提供成像场景中地物在可见光和近红外波长范围内的空间信息和地物属性诊断的光谱特征信息,在目标检测与识别方面拥有得天独厚的天然优势。然而,基于高光谱图像数据的目标检测也存在一定缺陷,如经典的高光谱目标检测算法仅利用光谱维度信息检测目标,检测模型要么对背景高维特征矩阵构建的准确度不足,要么对背景先验光谱特征的完备性要求较高,导致算法对不同复杂度的检测场景适应性不强。因此,基于计算复杂度较低、参数需求量较少且检测性能较为优异的经典多目标检测算法—多目标约束能量最小化(MCEM),提出了一种基于目标与背景环境特征分离模型的高光谱目标检测修正算法(R-MCEM)。首先,设计了一个与目标形状、尺寸相近的逐像元移动运算窗口,依次计算窗口中的每个像元与窗口内其他像元的光谱距离之和D1,像元与各类目标的光谱距离之和D2。其次,采用获得D1/D2最小值的像元替换窗口内的所有像元值。然后,自左向右、自上而下逐像元移动窗口,重复窗口内每一个像元与目标、背景像元的光谱距离运算,并确定窗口内与背景相似度最高、与目标相似度最低的像元。直到移动运算窗口遍历整个高光谱图像,大幅提升了基于目标与背景...     

Accurate object detection with a discriminative shape model

Discriminative model over bag-of-visual-words representation significantly improves the accuracy of object detection under clutter. However, it encounters bottleneck because of completely ignoring geometric constraint between features. On the contrary, to detect object accurately explicit shape model heavily relies on geometric information of the object, which as a result lacks of discriminative power. In this paper, we present a discriminative shape model to make use of the advantages of the two models based on the insight that the two models are essentially complementary. Discriminative model provides discriminative power, while shape model encodes geometry. The cost function that we used to distinguish objects considers both the detection maps of the discriminative model and the result of shape matching. In this cost function, we adopt a novel way to deal with multi-scale detection maps. We show that this cost function has very strong discriminative power, which makes learning a discriminative threshold for full object detection possible. For shape model, we also present a scheme for learning a good shape model from noisy images. Experiments on UIUC Car and Weizmann–Shotton horses show state-of-the-art performance of our model.     

Robust object detection based on deformable part model and improved scale invariant feature transform

We propose an approach to improve the detection results of a generic offline trained detector on frames from a specific video. For two consecutive frames of a video with the object, deformable part model (DPM) detection is performed to get the original detections. Then the image patches corresponding to the detected root box and part boxes were respectively obtained. Thirdly, improved scale invariant feature transform features (SIFT) from those image patches were extracted and matched with the SIFT features by KD-Tree. K-means clustering the angle and scale of matched keypoints to filter out the uncorrected matches and further remove false matches by RANSAC algorithm. Finally, the SIFT_DPM detection result from the matches between image patches of continuous frames was obtained. We focus on methods with high precision detection results since it is necessitated in real application. Extensive experiments with state-of-the-art detector demonstrate the efficacy of our approach.     

Small object detection in forward-looking infrared images with sea clutter using context-driven Bayesian saliency model

There are two common challenges for small object detection in forward-looking infrared (FLIR) images with sea clutter, namely, detection ambiguity and scale variance. This paper presents a context-driven Bayesian saliency model to deal with these two issues. By inspecting the camera geometry of the FLIR imaging under the background of sea and sky, we observed that there exists dependency relationship between the locations and scales at which objects may occur, and the context which is defined to be the location of horizon line. Based on this observation, we propose to incorporate contextual information into the basic bottom-up saliency computation, and a unified Bayesian model is developed to achieve this goal. The proposed model is generic and can be potentially applied to other circumstances where context is available for facilitating object detection. Experimental results have demonstrated the effectiveness of our method.     

基于局部尺度不变特征的快速目标识别

介绍了图像局部尺度不变特征的提取方法,将局部尺度不变特征用于目标识别,为提高识别实时性,提出利用金字塔和尺度空间的混合多尺度表示方法,按照从大尺度到小尺度的顺序对待识别图像的特征点进行检测与匹配,直到完成识别为止,有效地提高了识别速度。     

无人机视频运动目标实时检测及跟踪

对无人机拍摄视频中的地面运动目标提出一种实时检测跟踪算法。该算法利用特征点的对应关系将图像对配准,再对配准图像进行变化检测,根据变化和运动信息检测目标并消除虚警,将检测与跟踪相结合,对目标跟踪失效的情况,能重新正确定位目标,从而获取目标的完整运动轨迹。采用无人机拍摄的地面车辆图像的测试结果表明,算法能有效检测跟踪运动目标,并能达到25 f/s以上的实时处理速度。     

Contour level object detection with top-down information

This paper presents a contour level object detection approach. In contrast to conventional bounding box results, we give out the salient closed contour of the object, which provides a possibility of semantic analysis for the object. We get the salient closed contour with Ratio Contour algorithm. The top-down information needed by salient closed contour extraction is based on the well-known Bag-of-Features methodology. Our top-down information based contour extraction and completion is much more efficient and robust than many related approaches lack of the top-down information. We also propose a novel post-processing framework for object detection. With low threshold and a refined binary classifier, we can get stable high performance. We evaluate our approaches on UIUC cars dataset. We show that our approaches apparently improve the performance of object detections under clutter.     

Recognition of external object features in gas media using ultrasound transmission tomography

The paper presents and analyzes a new way of recognizing external object features (shape, size, location) in gas media using ultrasound transmission tomography (UTT) with parallel-ray-projection scanning geometry. The concept of UTT in a gas medium is close to classical tomography, however because of the nature of the measurement environment, visualizing the internal structure of solid objects is difficult; whereas it is possible to image their external features: their shape, size and spatial location. The paper presents the results of examining the shape, size and location of different objects in the air in the form of tomographic images, obtained in parallel-ray-projection geometry, using a specially elaborated research setup for UTT. Applying parallel-ray-projection geometry enabled us to investigate the influence of scanning resolution on image quality. In order to test the operation of the elaborated algorithm of tomographic image reconstruction on the basis of correct measurement data, special software was written for simulating the binary matrix of the measurements for a set of a dozen or so solid objects of different shapes and a few simulations were performed.     

Wood species identification using improved active shape model

In this paper, we propose a robust wood species identification scheme by using color wood surface images. First, a novel wood image acquirement system is devised, and the wood color image is converted into a V₁V₂I color-base image. Second, the corresponding grey histograms for V₁ and V₂ are established. Third, an improved active shape model is used to fulfill the curve deformation of the histogram curve of the standard specimen. This active shape model will then converge to the histogram curve of the test specimen. Finally, wood recognition is performed by comparing the initial and final active shape models with the histogram curve of the test specimen. We have experimentally proved that this scheme improves the mean recognition accuracy to approximately 90% for 5 wood species and that it can also be applied to the Gaussian noisy images. Moreover, the recognition accuracy can be further improved by combining this scheme with the texture feature recognition.