Visual Background Research Articles

Motion detection and mismatching correction by fusing laser and vision

Aiming at the limitations of single sensor in dynamic scene perception issue, an implementation system for fusing laser and vision was designed. In addition, two improved algorithms were proposed to solve the problems of the error foreground detection in the motion detection and the mismatching between the point clouds of different sensors. As for motion detection, the laser foreground points were firstly detected based on visual background subtraction algorithm. Then, the visual foreground was clustered regarding laser foreground points as the heuristic information. To solve the mismatching of fusion, the laser and vision point cloud were segmented into clusters based on the cell mismatching degree firstly. Then the corresponding stereo point cloud was registered referring to laser clusters. The corrected point cloud could be used for further verification by reconstructing the scene after filtering. The experimental results showed that the fusion foreground obtained finally had a better robustness to shadow. Compared with the whole registration correction, the average mismatching degree reduced by 75%, and the positive ratio in the direction of y and z converged at least 5%.

Visual Sensory Signals Dominate Tactile Cues during Docked Feeding in Hummingbirds.

Animals living in and interacting with natural environments must monitor and respond to changing conditions and unpredictable situations. Using information from multiple sensory systems allows them to modify their behavior in response to their dynamic environment but also creates the challenge of integrating different, and potentially contradictory, sources of information for behavior control. Understanding how multiple information streams are integrated to produce flexible and reliable behavior is key to understanding how behavior is controlled in natural settings. Natural settings are rarely still, which challenges animals that require precise body position control, like hummingbirds, which hover while feeding from flowers. Tactile feedback, available only once the hummingbird is docked at the flower, could provide additional information to help maintain its position at the flower. To investigate the role of tactile information for hovering control during feeding, we first asked whether hummingbirds physically interact with a feeder once docked. We quantified physical interactions between docked hummingbirds and a feeder placed in front of a stationary background pattern. Force sensors on the feeder measured a complex time course of loading that reflects the wingbeat frequency and bill movement of feeding hummingbirds, and suggests that they sometimes push against the feeder with their bill. Next, we asked whether the measured tactile interactions were used by feeding hummingbirds to maintain position relative to the feeder. We created two experimental scenarios—one in which the feeder was stationary and the visual background moved and the other where the feeder moved laterally in front of a white background. When the visual background pattern moved, docked hummingbirds pushed significantly harder in the direction of horizontal visual motion. When the feeder moved, and the background was stationary, hummingbirds generated aerodynamic force in the opposite direction of the feeder motion. These results suggest that docked hummingbirds are using visual information about the environment to maintain body position and orientation, and not actively tracking the motion of the feeder. The absence of flower tracking behavior in hummingbirds contrasts with the behavior of hawkmoths, and provides evidence that they rely primarily on the visual background rather than flower-based cues while feeding.

The effect of visual cues on performance in the ultimatum game in individuals with autism spectrum disorder

The Ultimatum Game (UG) allows for the assessment of altruistic behavior as well as the perception of fairness. We examined the effects of visual social cues (gaze of others), factors associated with autism, and trust on UG performance in typical adults (TAs) and individuals with autism spectrum disorder (ASD). We hypothesized that individuals with ASD would be less affected by visual social cues than TAs. We recruited 30 TAs and 30 individuals with ASD. Participants completed 30 trials of the UG, during which the visual background was altered to include either stylized eyespots, flowers, or a neutral background. Reaction times and money distributed in each condition were recorded. Reaction times did not vary among background conditions in either group, although individuals with ASD responded more slowly overall. TAs distributed less money in the neutral background and flowers conditions than in the eyespots condition, while no significant differences in the amount of money distributed were observed among background conditions for individuals with ASD, who also distributed more money overall than TAs. Such findings may be due to decreased susceptibility to social cues among individuals with ASD.

Extraction technique of region of interest from stereoscopic video

A feature fusion approach is presented to extract the region of interest (ROI) from the stereoscopic video. [0]Based on human vision system (HVS), the depth feature, the color feature and the motion feature are chosen as vision features. [0]The algorithm is shown as follows. Firstly, color saliency is calculated on superpixel scale. Color space distribution of the superpixel and the color difference between the superpixel and background pixel are used to describe color saliency and color salient region is detected. Then, the classic visual background extractor (Vibe) algorithm is improved from the update interval and update region of background model. The update interval is adjusted according to the image content. The update region is determined through non-obvious movement region and background point detection. So the motion region of stereoscopic video is extracted using improved Vibe algorithm. The depth salient region is detected by selecting the region with the highest gray value. Finally, three regions are fused into final ROI. Experiment results show that the proposed method can extract ROI from stereoscopic video effectively. In order to further verify the proposed method, stereoscopic video coding application is also carried out on the joint model (JM) encoder with different bit allocation in ROI and the background region.

Shooting for Smarter Motion Detection in Cameras: Improvements for the Visual Background Extractor Algorithm Using Optical Flow

Motion detection is one of the hot-and difficult-topics in the field of computer vision and image processing. The visual background extractor (ViBe) algorithm has a high calculation speed and a simple implementation, which is suitable for embedded systems. However, it still has some defects with respect to ghost suppression and dynamic background interferences. In this article, a new motion detection method called optical flow ViBe (OF-ViBe) is proposed, which combines the improved ViBe algorithm and motion information. We not only introduce the speed information for removing the ghost and noise points more accurately but also reduce the dynamic background interferences using direction information. The detection results in different videos demonstrate that our method improves the accuracy and robustness of the ViBe algorithm with little additional time cost.

Behavioral responses of Frankliniella occidentalis to floral volatiles combined with different background visual cues

The Western flower thrips, Frankliniella occidentalis (Thysanoptera: Thripidae), is one of the most destructive sucking pests of flowering plants. We examined behavioral responses of F. occidentalis adults to the floral volatiles of Rosa chinensis, Gardenia jasminoides, and Tagetes erecta in a Y-tube olfactometer with background visual cues (green, red, white, yellow, and blue backgrounds). The results show that F. occidentalis adults had similar responses under all the different background colors, but the responses by sex varied significantly when offered floral volatile or clean air. In pairings of floral volatiles, female F. occidentalis adults presented significant preferences (R. chinensis > G. jasminoides > T. erecta) in each treatment, while male F. occidentalis adults had no significant preference. Furthermore, the responses of female F. occidentalis adults to the three different floral volatiles compared with each other were greater under green and blue backgrounds than red, white, or yellow backgrounds. However, the responses of male F. occidentalis adults to floral volatiles were similar to all five backgrounds. These results indicate that female thrips could discriminate different floral volatiles, while male thrips could not. Both male and female thrips performed steady preferences of floral volatiles under all the five visual backgrounds, and the behavioral responses of female thrips to the floral volatiles could be enhanced under the green and blue backgrounds. These findings could hopefully aid in the development of effective trapping and monitoring strategies for this pest.

Aquatic prey use countershading camouflage to match the visual background

Lay SummaryAnimals often have darker coloration on their top surface than on their underside (“countershading”). In terrestrial prey, this patterning eliminates the body’s shadows to facilitate camouflage. In a freshwater fish that can change color, countershading provides optimal camouflage for different visual backgrounds and viewing angles rather than to reduce shadowing. Countershading camouflage may operate differently in terrestrial and aquatic habitats.

California scrub-jays reduce visual cues available to potential pilferers by matching food colour to caching substrate.

Some animals hide food to consume later; however, these caches are susceptible to theft by conspecifics and heterospecifics. Caching animals can use protective strategies to minimize sensory cues available to potential pilferers, such as caching in shaded areas and in quiet substrate. Background matching (where object patterning matches the visual background) is commonly seen in prey animals to reduce conspicuousness, and caching animals may also use this tactic to hide caches, for example, by hiding coloured food in a similar coloured substrate. We tested whether California scrub-jays (Aphelocoma californica) camouflage their food in this way by offering them caching substrates that either matched or did not match the colour of food available for caching. We also determined whether this caching behaviour was sensitive to social context by allowing the birds to cache when a conspecific potential pilferer could be both heard and seen (acoustic and visual cues present), or unseen (acoustic cues only). When caching events could be both heard and seen by a potential pilferer, birds cached randomly in matching and non-matching substrates. However, they preferentially hid food in the substrate that matched the food colour when only acoustic cues were present. This is a novel cache protection strategy that also appears to be sensitive to social context. We conclude that studies of cache protection strategies should consider the perceptual capabilities of the cacher and potential pilferers.

Improving Video Segmentation by Fusing Depth Cues and the Visual Background Extractor (ViBe) Algorithm.

Depth-sensing technology has led to broad applications of inexpensive depth cameras that can capture human motion and scenes in three-dimensional space. Background subtraction algorithms can be improved by fusing color and depth cues, thereby allowing many issues encountered in classical color segmentation to be solved. In this paper, we propose a new fusion method that combines depth and color information for foreground segmentation based on an advanced color-based algorithm. First, a background model and a depth model are developed. Then, based on these models, we propose a new updating strategy that can eliminate ghosting and black shadows almost completely. Extensive experiments have been performed to compare the proposed algorithm with other, conventional RGB-D (Red-Green-Blue and Depth) algorithms. The experimental results suggest that our method extracts foregrounds with higher effectiveness and efficiency.

Active background choice facilitates crypsis in a tropical crab

AbstractAnimals can evade predators in multiple ways, one of the most effective of which is to avoid detection in the first place. We know much about the evolution of color patterns that match the visual background to avoid detection (i.e., crypsis), yet we know surprisingly less about the specific behaviors that have co‐evolved with these morphological traits to enhance or maintain crypsis. We here explore whether the match between body color and background in a seemingly well‐camouflaged tropical shore crab is a result of active background choice. Taking advantage of a coastal area in the Solomon Islands with variable sand color and a population of the pallid ghost crab Ocypode pallidula with varying carapace color, we experimentally tested whether individuals actively choose specific substrate that best matches their color patterns. We found that individuals taken from extreme sand colors chose substrate that maintained crypsis, with relatively darker crabs typically choosing dark sand and lighter crabs choosing light sand. Crabs of intermediate color pattern, in contrast, showed no clear preference for dark or light sand. Our results suggest that potential prey can actively choose specific backgrounds to enhance and maintain crypsis, providing insights into how behavior interacts with morphological traits to avoid predator detection.

Spatial-temporal texture features for 3D human activity recognition using laser-based RGB-D videos

The IR camera and laser-based IR projector provide an effective solution for real-time collection of moving targets in RGB-D videos. Different from the traditional RGB videos, the captured depth videos are not affected by the illumination variation. In this paper, we propose a novel feature extraction framework to describe human activities based on the above optical video capturing method, namely spatial-temporal texture features for 3D human activity recognition. Spatial-temporal texture feature with depth information is insensitive to illumination and occlusions, and efficient for fine-motion description. The framework of our proposed algorithm begins with video acquisition based on laser projection, video preprocessing with visual background extraction and obtains spatial-temporal key images. Then, the texture features encoded from key images are used to generate discriminative features for human activity information. The experimental results based on the different databases and practical scenarios demonstrate the effectiveness of our proposed algorithm for the large-scale data sets.

Geographic divergence and colour change in response to visual backgrounds and illumination intensity in bearded dragons.

Animals may improve camouflage by both dynamic colour change and local evolutionary adaptation of colour but we have little understanding of their relative importance in colour-changing species. We tested for differences in colour change in response to background colour and light intensity in two populations of central bearded dragon lizards (Pogona vitticeps) representing the extremes in body coloration and geographical range. We found that bearded dragons change colour in response to various backgrounds and that colour change is affected by illumination intensity. Within-individual colour change was similar in magnitude in the two populations but varied between backgrounds. However, at the endpoints of colour change, each population showed greater similarity to backgrounds that were representative of the local habitat compared with the other population, indicating local adaptation to visual backgrounds. Our results suggest that even in species that change colour, both phenotypic plasticity and geographic divergence of coloration may contribute to improved camouflage.

Improved visual background extractor with adaptive range change

The visual background extractor (ViBe) has become one of the best motion object detection algorithms because of its good detection results and low memory requirements. However, the ViBe model cannot self-adjust the value range of the parameter that controls the number of samples chosen from the background template. In this paper, two models are proposed to help automatically change the parameter range in different environments. The blink energy model can detect dynamic backgrounds by increasing the range, while the object probability model can prevent corrosion of motion objects by decreasing the range. The experimental results show that our proposed method can both accurately recognize dynamic backgrounds and efficiently prevent object corrosion. In addition, our method shows better performance on benchmark datasets than several commonly used detection algorithms.

Vertical head translation impairs dynamic visual acuity during near viewing.

Dynamic visual acuity is an important clinical tool for assessment of the rotational vestibulo-ocular reflex (rVOR). It is based on the fact that the normal rVOR stabilizes vision and maintains visual acuity during head rotation. The translational VOR (tVOR) generates eye movements during linear head motion. Studies in humans have shown that gaze stabilization during translation is incomplete and that there is a strong effect of the visual environment: eye velocity is much greater in the light than in the dark. In this study, we measured visual acuity during vertical translation in 11 subjects and asked whether a more complex visual background would enhance the response and improve acuity. During 2 Hz whole-body translation, tumbling-E optotypes (0.0-0.9 logMAR in steps of 0.1 logMAR, six trials of each size randomly ordered) were flashed on a screen that was 30 cm in front of the subject's eyes. The subject reported the optotype's orientation with a joystick. Based on a threshold of 75% trials correctly identified, the group dynamic acuity was 0.72 logMAR, compared to a static acuity of 0.0 logMAR. When the background was enhanced with a stationary dot pattern, dynamic acuity improved to 0.42 logMAR. Our findings show that vertical head translation degrades vision more than head rotation. This may limit the use of dynamic acuity as a clinical measure of otolith function.

Size Matters: Observed and Modeled Camouflage Response of European Cuttlefish (Sepia officinalis) to Different Substrate Patch Sizes during Movement.

Camouflage is common throughout the phylogenetic tree and is largely used to minimize detection by predator or prey. Cephalopods, and in particular Sepia officinalis cuttlefish, are common models for camouflage studies. Predator avoidance behavior is particularly important in this group of soft-bodied animals that lack significant physical defenses. While previous studies have suggested that immobile cephalopods selectively camouflage to objects in their immediate surroundings, the camouflage characteristics of cuttlefish during movement are largely unknown. In a heterogenic environment, the visual background and substrate feature changes quickly as the animal swim across it, wherein substrate patch is a distinctive and high contrast patch of substrate in the animal's trajectory. In the current study, we examine the effect of substrate patch size on cuttlefish camouflage, and specifically the minimal size of an object for eliciting intensity matching response while moving. Our results indicated that substrate patch size has a positive effect on animal's reflectance change, and that the threshold patch size resulting in camouflage response falls between 10 and 19 cm (width). These observations suggest that the animal's length (7.2–12.3 cm mantle length in our case) serves as a possible threshold filter below which objects are considered irrelevant for camouflage, reducing the frequency of reflectance changes—which may lead to detection. Accordingly, we have constructed a computational model capturing the main features of the observed camouflaging behavior, provided for cephalopod camouflage during movement.

The foreground detection algorithm combined the temporal–spatial information and adaptive visual background extraction

ABSTRACTVisual background extraction algorithm, which utilises a global threshold to complete the foreground segmentation, cannot adapt to illumination change well. It will easily choose the wrong pixels to initialise the background model, resulting in the emergence of the ghost in the beginning of detection. In order to address these problems, this article proposes an improved algorithm based on pixel’s temporal–spatial information to initialise the background model. First of all, the pixels in video image sequences and their neighbourhood pixels are used to complete background model initialisation in the first five frames. Second, the segmentation threshold is adaptively obtained by the complexity of background that uses the spatial neighbourhood pixels. Finally, the background model of the neighbourhood pixels is updated by a dynamic update rate which is gained by calculating the Euclidean distance between pixels. Experimental results and comparative study illustrate that the improved method can not only increase the accuracy of target detection by reducing the impact of illumination change effectively but also eliminate the ghost quickly.

Research on visual background extractor to identify the vehicles based on edge similarity

By combining visual background extractor with Canny operator, a new approach to identify vehicles in a complex traffic environment is proposed. Firstly, the foreground object is extracted by ViBe algorithm and background difference algorithm, then the 'ghost' is removed by means of edge similarity. Next, the complete moving objects can be obtained by using morphological processing for the foreground object, which can be used to detect vehicles by combining with motion analysis. Experiments showed that the whole region of the vehicle object in a complex traffic environment can be extracted exactly and effectively by using this method. In addition, problems due ghosts and the variation of background can be well handled with low computation complexity, which can fulfil the needs of real-time operation.

Research on visual background extractor to identify the vehicles based on edge similarity

By combining visual background extractor with Canny operator, a new approach to identify vehicles in a complex traffic environment is proposed. Firstly, the foreground object is extracted by ViBe algorithm and background difference algorithm, then the 'ghost' is removed by means of edge similarity. Next, the complete moving objects can be obtained by using morphological processing for the foreground object, which can be used to detect vehicles by combining with motion analysis. Experiments showed that the whole region of the vehicle object in a complex traffic environment can be extracted exactly and effectively by using this method. In addition, problems due ghosts and the variation of background can be well handled with low computation complexity, which can fulfil the needs of real-time operation.

Examining Effects of Physical Exertion on the Dynamic Visual Acuity Test in Collegiate Athletes.

Acute symptoms of dizziness and/or imbalance commonly experienced in athletes postconcussion are speculated to arise from dysfunction at multiple levels (i.e., inner ear or central vestibular system) to appropriately integrate afferent sensory information. Disruption along any pathway of the balance system can result in symptoms of dizziness, decreased postural control function (vestibulospinal reflex), and reduced vestibulo-ocular reflex function. This may also lead to decreased gaze stability with movements of the head and may account for symptoms of blurred vision or diplopia reported in almost half of athletes sustaining a concussion. Current concussion position statements include measures of postural control to examine changes to the balance system postconcussion. The Balance Error Scoring System (BESS) is a commonly used low-cost postural control measure for concussion assessment. Although this is a widely used measure for documenting balance function on both immediate (sideline) and recovery monitoring, the BESS has been shown to be affected by physical exertion. Therefore, the BESS may not be the most efficient means of examining functional changes to the balance system immediately after head injury. Dynamic Visual Acuity Test (DVAT) has been found to effectively evaluate and monitor changes to the gaze stability system postinjury. Thus, DVAT may be an additional measure in the concussion assessment battery, as well as an alternative for more immediate sideline assessment to help make objective return-to-play decisions. The aim of the study was to determine the effects of physical exertion on a clinical vestibular assessment, the DVAT, in collegiate athletes, as a first step in defining the role of this measure in the concussion assessment battery. Cross-sectional, repeated-measures design. Twenty-eight healthy collegiate athletes (20 males, 8 females; age = 20.25 ± 1.46 yr, range = 18-25 yr) volunteered to participate in the study. Participants were randomly assigned to complete a 20-min protocol of physical exertion or rest. DVAT was completed pre-exertion or rest (pre-DVAT), immediately following the 20-min protocol (post-DVAT I), and again 10 min after the completion of the 20-min protocol (post-DVAT II). Ratings of perceived exertion (RPE) and heart rate (HR) were monitored throughout testing. Repeated-measures analysis of the variance were used to examine the effects of exertion on DVAT. Additionally, intraclass correlation coefficients were used to examine test reliability. No significant main effect was observed for right and left DVAT logarithm of the minimal angle of resolution loss between groups or across time points (p > 0.05). A significant main effect was observed for RPE and HR for groups and time points (p < 0.001), indicating adequate physical exertion and rest. Fair to good reliability (intraclass correlation coefficient values between 0.4 and 0.74) was observed for both rightward and leftward movements of the head across the three time points. Findings from this study suggest that DVAT is not affected by physical exertion and may provide a more immediate assessment of the balance system that may be of use for the sideline concussion assessment. Future studies will be performed to examine additional factors (e.g., background noise, complex visual backgrounds) that may affect DVAT performance in the sideline environment.

Scaffold analysis of PubChem database as background for hierarchical scaffold-based visualization

BackgroundVisualization of large molecular datasets is a challenging yet important topic utilised in diverse fields of chemistry ranging from material engineering to drug design. Especially in drug design, modern methods of high-throughput screening generate large amounts of molecular data that call for methods enabling their analysis. One such method is classification of compounds based on their molecular scaffolds, a concept widely used by medicinal chemists to group molecules of similar properties. This classification can then be utilized for intuitive visualization of compounds.ResultsIn this paper, we propose a scaffold hierarchy as a result of large-scale analysis of the PubChem Compound database. The analysis not only provided insights into scaffold diversity of the PubChem Compound database, but also enables scaffold-based hierarchical visualization of user compound data sets on the background of empirical chemical space, as defined by the PubChem data, or on the background of any other user-defined data set. The visualization is performed by a web based client-server application called Scaffvis. It provides an interactive zoomable tree map visualization of data sets up to hundreds of thousands molecules. Scaffvis is free to use and its source codes have been published under an open source license.Graphical abstract. Electronic supplementary materialThe online version of this article (doi:10.1186/s13321-016-0186-7) contains supplementary material, which is available to authorized users.