Neural synchrony correlates with surface segregation rules

To analyse an image, the visual system must decompose the scene into its relevant parts. Identifying distinct surfaces is a basic operation in such analysis, and is believed to precede object recognition. Two superimposed gratings moving in different directions (plaid stimuli) may be perceived either as two surfaces, one being transparent and sliding on top of the other (component motion) or as a single pattern whose direction of motion is intermediate to the component vectors (pattern motion). The degree of transparency, and hence the perception, can be manipulated by changing only the luminance of the grating intersections. Here we show that neurons in two visual cortical areas--A18 and PMLS--synchronize their discharges when responding to contours of the same surface but not when responding to contours belonging to different surfaces. The amplitudes of responses correspond to previously described rate predictions for component and pattern motion, but, in contrast to synchrony, failed to reflect the transition from component to pattern motion induced by manipulating the degree of transparency. Thus, dynamic changes in synchronization could encode, in a context-dependent way, relations among simultaneous responses to spatially superimposed contours and thereby bias their association with distinct surfaces.     

Neuronal synchrony does not correlate with motion coherence in cortical area MT

Natural visual scenes are cluttered with multiple objects whose individual features must somehow be selectively linked (or 'bound') if perception is to coincide with reality. Recent neurophysiological evidence supports a 'binding-by-synchrony' hypothesis: neurons excited by features of the same object fire synchronously, while neurons excited by features of different objects do not. Moving plaid patterns offer a straightforward means to test this idea. By appropriate manipulations of apparent transparency, the component gratings of a plaid pattern can be seen as parts of a single coherently moving surface or as two non-coherently moving surfaces. We examined directional tuning and synchrony of area-MT neurons in awake, fixating primates in response to perceptually coherent and non-coherent plaid patterns. Here we show that directional tuning correlated highly with perceptual coherence, which is consistent with an earlier study. Although we found stimulus-dependent synchrony, coherent plaids elicited significantly less synchrony than did non-coherent plaids. Our data therefore do not support the binding-by-synchrony hypothesis as applied to this class of motion stimuli in area MT.     

Parallel processing of motion and colour information

When the two eyes are confronted with sufficiently different versions of the visual environment, one or the other eye dominates perception in alternation. A similar situation may be created in the laboratory by presenting images to the left and right eyes which differ in orientation or colour. Although perception is dominated by one eye during rivalry, there are a number of instances in which visual processes nevertheless continue to integrate information from the suppressed eye. For example the interocular transfer of the motion after-effect is undiminished when induced during binocular rivalry. Thus motion information processing may occur in parallel with the rivalry process. Here we describe a novel example in which the visual system simultaneously exhibits binocular rivalry and vision that integrates signals from both eyes. This apparent contradiction is resolved by postulating parallel visual processes devoted to the analyses of colour and motion information. Counterphased gratings are viewed dichoptically such that for one eye the grating is composed of alternating yellow and black stripes (luminance) while for the other it is composed of alternating red and green stripes (chrominance). When the gratings are fused, a moving grating is perceived. A consistent direction of motion can only be achieved if left and right monocular signals are integrated by the nervous system. Yet the apparent colour of the binocular percept alternates between red-green and yellow-black. These observations demonstrate the segregation of processing by the early motion system from that affording the perception of colour. Although, in this stimulus, colour information in itself can play no part in the cyclopean perception of motion direction, colour is carried along perceptually (filled in) by the moving pattern which is integrated from both eyes.     

The perception of moving plaids reveals two motion-processing stages

When viewed through a small aperture, the perceived motion exhibited by a long moving line or grating is ambiguous. This situation prevails because even a perfect machine could only detect motion perpendicular to a moving contour, so motion parallel to a contour is undetectable. The human visual system views the world through an aperture array--the neural receptive fields. Therefore a moving object is viewed through many small apertures and the motion within many of those apertures is ambiguous. This ambiguity may be resolved by monitoring the motion of a distinctive feature, such as a line-end or corner, and attributing to the larger object the motion of the feature. Alternatively, Adelson and Movshon have suggested that moving images are processed in two stages, that is, they are first decomposed into one-dimensional components which are later recombined to generate perceived object motion. For a moving plaid, defined as the sum of two drifting gratings, these alternative models generate different predictions concerning the resolution of the plaid's motion ambiguity. A feature monitor would respond to the motion of the intersections between gratings, whereas the two-stage motion processor would first decompose the plaid into its constituent gratings and subsequently recombine them to generate the perception of a moving plaid. Using speed discrimination to distinguish between the two models, I find that discrimination thresholds reflect the speed of a plaid's component gratings, rather than the speed of the plaid itself. This result supports the two-stage model. Although speed discrimination is limited by component processing, observers cannot directly access component speed. The only perceptually accessible velocity signal is generated by the second-stage pattern processing.     

Neural correlates of perceptual motion coherence.

The motions of overlapping contours in a visual scene may arise from the physical motion(s) of either a single or multiple surface(s). A central problem facing the visual motion system is that of assigning the most likely interpretation. The rules underlying this perceptual decision can be explored using a visual stimulus formed by superimposing two moving gratings. The resultant percept is either that of a single coherently moving 'plaid pattern' (coherent motion) or of the two component gratings sliding noncoherently across one another (noncoherent motion). When plaid patterns are configured to mimic one transparent grating overlying another, the percept of noncoherent motion dominates. We now report that neurons in the visual cortex of rhesus monkeys exhibit changes in direction tuning that parallel this perceptual phenomenon: sensitivity to the motions of the component gratings is enhanced under conditions that favour the perception of noncoherent motion. These results challenge models of cortical visual processing that fail to take into account the contribution of figural image segmentation cues to the analysis of visual motion.     

动态场景中运动目标检测与跟踪

为了在静态和动态场景中均能实现对运动目标的检测与跟踪,提出了基于运动检测和视频跟踪相结合的视频监控方法. 建立四参数运动仿射模型来描述全局运动,采用块匹配法对其进行参数估计;采用基于全局运动补偿的Horn-Schunck算法检测出运动目标;使用卡尔曼滤波对运动目标的质心位置、宽度和高度进行跟踪. 实验结果表明,该方法能够有效地对静态和动态场景中运动目标进行检测与跟踪.     

实时视频对象识别与计数系统的模型和算法设计

动态视频对象的识别与检测是计算机图像处理领域中的一个前沿课题。该文提出了一种利用实时视频图像识别技术 ,对无规则行进队伍的人员进行检测与统计的系统模型和算法设计。基于颜色模型转换、减影、膨胀、腐蚀、聚类、匹配、跟踪等技术 ,实现了对特定检测区域内运动对象的识别、检测和计数。着重讨论了系统的设计原理和实现方法 ,以及对象运动的物理模型和多目标识别跟踪算法。该系统于1998年开始成功地应用于毛主席纪念堂 ,对每日的瞻仰人数进行实时统计 ,计数误差小于 3% ,系统工作稳定可靠     

Motion-Compensated Reconstruction Method Based on Rigid Motion Model with Multi-Object

This paper gives an efficient approach to reconstruct moving multiple objects(multi-object).Each object has independently rigid motion which includes translation and rotation.The traditional FBP algorithm can resolve the one-object motion problem rather well.However,it suffers from perceptible motion artifacts in multi-object cases.This paper proposes a new motion-compensated reconstruction approach with a priori knowledge of the rigid motion model.Both an FBP-type and an ART-type algorithm were derived.In ...     

Motion-deblurring in human vision

If photographs are taken of moving objects at slow shutter speeds the images of the objects are blurred. In human vision, however, we are not normally conscious of blur from moving objects despite the fact that the temporal response of the photoreceptors is sluggish. It has been suggested that there are motion-deblurring mechanisms specifically to aid the visual system in the analysis of the shape of retinally moving targets. Models of motion deblurring have been influenced by the finding that certain very precise spatial pattern discriminations are unaffected by motion. An example is vernier hyperacuity, in which the observer must detect the direction of offset between two lines with abutting ends. With a stationary stimulus, observers can detect a vernier cue of less than 10 arcsec and acuity is unaffected by retinal-image motion of up to 3 deg s-1 We confirm this finding, but provide evidence against any general deblurring mechanism by showing that another kind of hyperacuity, discrimination of the distance between two parallel lines (spatial interval acuity), is interfered with by motion. This argues against a general deblurring mechanism, such as a neural network 'shifter circuit', and we point out that the high level of vernier acuity for moving stimuli is susceptible to an alternative explanation.     

fMRI evidence for objects as the units of attentional selection.

Contrasting theories of visual attention emphasize selection by spatial location, visual features (such as motion or colour) or whole objects. Here we used functional magnetic resonance imaging (fMRI) to test key predictions of the object-based theory, which proposes that pre-attentive mechanisms segment the visual array into discrete objects, groups, or surfaces, which serve as targets for visual attention. Subjects viewed stimuli consisting of a face transparently superimposed on a house, with one moving and the other stationary. In different conditions, subjects attended to the face, the house or the motion. The magnetic resonance signal from each subject's fusiform face area, parahippocampal place area and area MT/MST provided a measure of the processing of faces, houses and visual motion, respectively. Although all three attributes occupied the same location, attending to one attribute of an object (such as the motion of a moving face) enhanced the neural representation not only of that attribute but also of the other attribute of the same object (for example, the face), compared with attributes of the other object (for example, the house). These results cannot be explained by models in which attention selects locations or features, and provide physiological evidence that whole objects are selected even when only one visual attribute is relevant.     

Motion Tracking with Fast Adaptive Background Subtraction

0　ＩｎｔｒｏｄｕｃｔｉｏｎＩｎａｎｉｎｔｅｌｌｉｇｅｎｔｖｉｄｅｏｓｕｒｖｅｉｌｌａｎｃｅａｎｄｍｏｎｉｔｏｒｉｎｇｓｙｓｔｅｍ ,ｏｎｅｏｆｔｈｅｍｏｓｔｃｏｍｐｌｅｘａｎｄｉｍｐｏｒｔａｎｔｔａｓｋｓｉｓｒｅａｌｔｉｍｅｅｘｔｒａｃｔｉｎｇａｎｄｔｒａｃｋｉｎｇｍｏｔｉｏｎｏｂｊｅｃｔｓ .Ｉｎｇｅｎｅｒａｌ,ｔｈｅｒｅｅｘｉｓｔｔｗｏｍｅｔｈｏｄｓｆｏｒｅｘｔｒａｃｔｉｏｎｎａｍｅｄａｓｆｏｒｅｇｒｏｕｎｄｅｘｔｒａｃｔｉｏｎａｎｄｂａｃｋｇｒｏｕｎｄｓｕｂｔｒａｃｔｉｏｎ ,ｒｅｓｐｅｃｔｉｖ…     

Spatial filtering precedes motion detection.

When we perceive motion on a television or cinema screen, there must be some process that allows us to track moving objects over time: if not, the result would be a conflicting mass of motion signals in all directions. A possible mechanism, suggested by studies of motion displacement in spatially random patterns, is that low-level motion detectors have a limited spatial range, which ensures that they tend to be stimulated over time by the same object. This model predicts that the direction of displacement of random patterns cannot be detected reliably above a critical absolute displacement value (Dmax) that is independent of the size or density of elements in the display. It has been inferred that Dmax is a measure of the size of motion detectors in the visual pathway. Other studies, however, have shown that Dmax increases with element size, in which case the most likely interpretation is that Dmax depends on the probability of false matches between pattern elements following a displacement. These conflicting accounts are reconciled here by showing that Dmax is indeed determined by the spacing between the elements in the pattern, but only after fine detail has been removed by a physiological prefiltering stage: the filter required to explain the data has a similar size to the receptive field of neurons in the primate magnocellular pathway. The model explains why Dmax can be increased by removing high spatial frequencies from random patterns, and simplifies our view of early motion detection.     

运动目标的快速检测、跟踪和判别

为完成自然环境中大范围的环境监控 ,实现了一个运动目标检测、跟踪和判别系统。该系统利用一个固定平台上的、有 3 60°旋转和一定俯仰的两自由度摄像机监视自然环境 ,利用 2 -D仿射模型和鲁棒参数估计的主运动分析得到背景运动参数 ,能够在短时间内完成 3 60°全景图的拼接 ,并能利用出格点检测和聚类自动检测、通过限制搜索范围的检测和维护运动目标缓冲池主动跟踪运动目标 ,还能按目标区域的周期性变化判别目标种类 (人或车辆 )。实验表明 ,系统能够实时可靠地检测、跟踪运动目标并完成判别 ,满足特定的监控要求。另外 ,该运动目标判别方法简单可靠 ,其结果可作为视频序列识别和检索的一项重要特征     

一种基于SIFT的遮挡目标跟踪算法

针对运动目标遮挡的难题,提出一种新的遮挡目标跟踪算法.采用三帧取均值进行背景建模,采用相邻帧差法和背景差分结合自动提取出运动目标,对单运动目标生成SIFI(scale invariant feature transform)向量.当运动目标处于遮挡状态时,将遮挡区域与单运动目标进行SIFT特征匹配,通过特征匹配点的坐标...     

一种基于SIFT的遮挡目标跟踪算法

针对运动目标遮挡的难题,提出一种新的遮挡目标跟踪算法。采用三帧取均值进行背景建模,采用相邻帧差法和背景差分结合自动提取出运动目标,对单运动目标生成SIFT(scale invariant feature transform)向量。当运动目标处于遮挡状态时,将遮挡区域与单运动目标进行SIFT特征匹配,通过特征匹配点的坐标,找出单运动目标在遮挡区域中的位置,并对SIFT特征匹配运用RANSAC算法进行优化,实现遮挡情况下目标的有效跟踪。实验表明,该算法能准确地跟踪处于遮挡中的目标,实现运动目标跟踪的连续性和稳定性。     

Multi-features Based Approach for Moving Shadow Detection

IntroductionAs the core of smart surveillance system, image analysistechnology has been used to many fields including home andcommunity surveillance, and transportation monitoring. Themain aim is to detect special moving objects, such as vehicle,walking person, etc. By computing object features as motion,shape, color, texture, centroid, and area, the events in thescene can be evaluated. The requirements of moving objectdetection is the tracking application should be robus…     

基于帧差法和边缘检测法的视频分割算法

针对目前常用的视频分割方法易受到噪声、亮度突变的影响,很难提取出完整的运动目标这一问题,提出一种基于帧差法和边缘检测法相结合的视频分割算法.该算法首先对连续三帧图像进行差分得到运动区域,然后对当前帧进行Kirsch边缘检测得到边缘图像,综合二者的检测结果得到更为精准的运动对象边缘.采用边缘连接算法完成对断裂边缘的连接,最后通过区域填充得到运动目标掩模图像,从而分割出完整的运动目标.实验表明,该算法可以实时有效地将运动物体从视频序列中自动地分割出来.     

高速运动物体的图象分析与研究

按照狭义相对论,一个运动物体看起来将沿运动方向缩短为原来的１倍－Ｌｏｒｅｎｔｚ收缩,由此而见,高速飞船中的乘客观看球形物体时,看到的会是一个椭球吗？本文给出了否定的结论,并将证明：正是由于Ｌｏｒｅｎｔｚ收缩,我们会看到高速运动的物体只不过转了一个角度,并没有发生畸变,这个结论试图澄清人们对狭义相对论中关于高速运行物体视视形象描述的某些“事实”。     

The remarkable inefficiency of word recognition

Do we recognize common objects by parts, or as wholes? Holistic recognition would be efficient, yet people detect a grating of light and dark stripes by parts. Thus efficiency falls as the number of stripes increases, in inverse proportion, as explained by probability summation among independent feature detectors. It is inefficient to detect correlated components independently. But gratings are uncommon artificial stimuli that may fail to tap the full power of visual object recognition. Familiar objects become special as people become expert at judging them, possibly because the processing becomes more holistic. Letters and words were designed to be easily recognized, and, through a lifetime of reading, our visual system presumably has adapted to do this as well as it possibly can. Here we show that in identifying familiar English words, even the five most common three-letter words, observers have the handicap predicted by recognition by parts: a word is unreadable unless its letters are separately identifiable. Efficiency is inversely proportional to word length, independent of how many possible words (5, 26 or thousands) the test word is drawn from. Human performance never exceeds that attainable by strictly letter- or feature-based models. Thus, everything seen is a pattern of features. Despite our virtuosity at recognizing patterns and our expertise from reading a billion letters, we never learn to see a word as a feature; our efficiency is limited by the bottleneck of having to rigorously and independently detect simple features.     

Motion-induced spatial conflict

Borders defined by small changes in brightness (luminance contrast) or by differences in colour (chromatic contrast) appear to move more slowly than those defined by strong luminance contrast. As spatial coding is influenced by motion, if placed in close proximity, the different types of moving border might appear to drift apart. Using this configuration, we show here that observers instead report a clear illusory spatial jitter of the low-luminance-contrast boundary. This visible interaction between motion and spatial-position coding occurred at a characteristic rate (approximately 22.3 Hz), although the stimulus motion was continuous and invariant. The jitter rate did not vary with the speed of movement. The jitter was not due to small involuntary movements of the eyes, because it only occurred at a specific point within the stimulus, the low-luminance-contrast boundary. These findings show that the human visual system contains a neural mechanism that periodically resolves the spatial conflict created by adjacent moving borders that have the same physical but different perceptual speeds.