Irrelevant Objects Research Articles

Building Detection Using Enhanced HOG–LBP Features and Region Refinement Processes

Building detection from two-dimensional high-resolution satellite images is a computer vision, photogrammetry, and remote sensing task that has arisen in the last decades with the advances in sensors technology and can be utilized in several applications that require the creation of urban maps or the study of urban changes. However, the variety of irrelevant objects that appear in an urban environment and resemble buildings, and the significant variations in the shape and generally the appearance of buildings render building detection a quite demanding task. As a result, automated methods that can robustly detect buildings in satellite images are necessary. To this end, we propose a building detection method that consists of two modules. The first module is a feature detector that extracts histograms of oriented gradients (HOG) and local binary patterns (LBP) from image regions. Using a novel approach, a support vector machine classifier is trained with the introduction of a special denoising distance measure for the computation of distances between HOG–LBP descriptors before their classification to the building or nonbuilding class. The second module consists of a set of region refinement processes that employs the output of the HOG–LBP detector in the form of detected rectangular image regions. Image segmentation is performed and a novel building recognition methodology is proposed to accurately identify building regions, while simultaneously discard false detections of the first module of the proposed method. We demonstrate that the proposed methodology can robustly detect buildings from satellite images and outperforms state-of-the-art building detection methods.

Task Context Overrules Object- and Category-Related Representational Content in the Human Parietal Cortex.

The dorsal, parietal visual stream is activated when seeing objects, but the exact nature of parietal object representations is still under discussion. Here we test 2 specific hypotheses. First, parietal cortex is biased to host some representations more than others, with a different bias compared with ventral areas. A prime example would be object action representations. Second, parietal cortex forms a general multiple-demand network with frontal areas, showing similar task effects and representational content compared with frontal areas. To differentiate between these hypotheses, we implemented a human neuroimaging study with a stimulus set that dissociates associated object action from object category while manipulating task context to be either action- or category-related. Representations in parietal as well as prefrontal areas represented task-relevant object properties (action representations in the action task), with no sign of the irrelevant object property (category representations in the action task). In contrast, irrelevant object properties were represented in ventral areas. These findings emphasize that human parietal cortex does not preferentially represent particular object properties irrespective of task, but together with frontal areas is part of a multiple-demand and content-rich cortical network representing task-relevant object properties.

Broad attention to multiple individual objects may facilitate change detection with complex auditory scenes.

Attention and other processing constraints limit the perception of objects in complex scenes, which has been studied extensively in the visual sense. We used a change deafness paradigm to examine how attention to particular objects helps and hurts the ability to notice changes within complex auditory scenes. In a counterbalanced design, we examined how cueing attention to particular objects affected performance in an auditory change-detection task through the use of valid or invalid cues and trials without cues (Experiment 1). We further examined how successful encoding predicted change-detection performance using an object-encoding task and we addressed whether performing the object-encoding task along with the change-detection task affected performance overall (Experiment 2). Participants had more error for invalid compared to valid and uncued trials, but this effect was reduced in Experiment 2 compared to Experiment 1. When the object-encoding task was present, listeners who completed the uncued condition first had less overall error than those who completed the cued condition first. All participants showed less change deafness when they successfully encoded change-relevant compared to irrelevant objects during valid and uncued trials. However, only participants who completed the uncued condition first also showed this effect during invalid cue trials, suggesting a broader scope of attention. These findings provide converging evidence that attention to change-relevant objects is crucial for successful detection of acoustic changes and that encouraging broad attention to multiple objects is the best way to reduce change deafness. (PsycINFO Database Record

The Neural Dynamics of Attentional Selection in Natural Scenes.

Efficient attentional selection is crucial in many everyday situations. For example, when driving a car, we need to quickly detect obstacles, such as pedestrians crossing the street, while ignoring irrelevant objects. How can humans efficiently perform such tasks, given the multitude of objects contained in real-world scenes? Here we used multivariate decoding of magnetoencephalogaphy data to characterize the neural underpinnings of attentional selection in natural scenes with high temporal precision. We show that brain activity quickly tracks the presence of objects in scenes, but crucially only for those objects that were immediately relevant for the participant. These results provide evidence for fast and efficient attentional selection that mediates the rapid detection of goal-relevant objects in real-world environments.

Impact of Peripheral Field Loss on the Execution of Natural Actions: A Study With Glaucomatous Patients and Normally Sighted People.

We investigated the visuomotor behavior of people with reduced peripheral field due to glaucoma while they accomplished natural actions. Twelve participants with glaucoma and 13 normally sighted controls were included. Participants were asked to accomplish a familiar sandwich-making task and a less familiar model-building task with a children's construction set while their eye movements were recorded. Both scene layouts contained task-relevant and task-irrelevant objects. There was no time constraint. Participants with glaucoma were slower to perform the task than were the normal observers, but the slower performance was confined to the unfamiliar model-building task. Patients and controls were equally efficient in the more familiar sandwich-making task. On initial exposure, before the first reaching movement was initiated, patients scanned the objects longer than did controls, particularly in the unfamiliar model-building task, and controls fixated irrelevant objects less than did patients. During the working phase fixations were on average longer for patients than for controls and patients made more saccades than did controls. Patients did not grasp more irrelevant objects compared with controls. The results provide evidence that, although slower than controls, patients with glaucoma were able to accomplish natural actions efficiently even when the task required discrimination of small structurally similar objects (nuts and screws in the model-building task). Their difficulties were reflected in longer fixation times and more head and eye movements compared with controls, presumably to compensate for lower visibility when objects fell in the part of their visual field where sensitivity was reduced.

The effect of tDCS on task relevant and irrelevant perceptual learning of complex objects

During perceptual learning the visual representations in the brain are altered, but the causal role of these changes has not yet been fully characterized. Here we investigated the effect of tDCS delivered on the lateral occipital (LO) cortex on task-relevant and task-irrelevant learning of complex objects. Participants completed three training sessions and one test session on separate days. They were trained in two tasks: one relevant to object recognition (object naming in a backward masking paradigm) and one where objects were presented in the background but not relevant for the task (orientation judgment). Visual input was equated as much as possible between the tasks: an object and an oriented red line were presented in each trial in both tasks. The only crucial difference between the tasks was the relevance of the stimuli: the object was relevant for the object naming task, while the oriented red line but not the object was relevant for the orientation judgment task. During training, half of the participants received anodal tDCS stimulation on LO. In the test session, participants were tested on their ability to name the trained objects, the irrelevant objects presented during the orientation judgment task, and a new set of objects they were not exposed to during training. Participants stimulated with anodal tDCS during training showed a larger improvement of performance for the trained objects compared to participants in the sham condition. No learning effect was found for the irrelevant objects or the new objects. To conclude, this study suggests that LO plays a causal role in relevant object learning. Mere exposure is not sufficient to train object recognition in our paradigm. Meeting abstract presented at VSS 2016

An Efficient Unsupervised Clustered Adaptive Antihub Technique for Outlier Detection in High Dimensional Data

Objective: The objective of this paper is to find the inconsistent objects in data which has high dimension through reduced computation time and increased accuracy. Methods: Hubness specifically Antihubs (points that rarely occur in k nearest neighbor lists) is the newly recognized concept for handling data which has high dimension. The advanced version of Antihub is Antihub2 which is for reconsidering the outlier score of a point obtained by the Antihub method. However, regarding computation time, Antihub2 runs slower. This paper institutes an approach called AdaptiveAntihub2Clust, which is a clustered Adaptive Antihub technique for unsupervised outlier detection to reduce computation time and to improve the accuracy. Findings: The results of an existing Antihub2 method is compared with the proposed method called AdaptiveAntihub2Clust. The experimental results elucidate that AdaptiveAntihub2Clust outperforms well than Antihub2 and also resolved that there is not only a substantial decrease in computation time but also progress in accuracy occurred while the newly built approach is practically used for finding outliers. Applications: The irrelevant objects may ascend due to numerous faults. Detection of such objects identifies the mistakes and fraud before they deteriorate with terrible significances and cleanses the data for further processing.

Of "what" and "where" in a natural search task: Active object handling supports object location memory beyond the object's identity.

Looking for as well as actively manipulating objects that are relevant to ongoing behavioral goals are intricate parts of natural behavior. It is, however, not clear to what degree these two forms of interaction with our visual environment differ with regard to their memory representations. In a real-world paradigm, we investigated if physically engaging with objects as part of a search task influences identity and position memory differently for task-relevant versus irrelevant objects. Participants equipped with a mobile eye tracker either searched for cued objects without object interaction (Find condition) or actively collected the objects they found (Handle condition). In the following free-recall task, identity memory was assessed, demonstrating superior memory for relevant compared to irrelevant objects, but no difference between the Handle and Find conditions. Subsequently, location memory was inferred via times to first fixation in a final object search task. Active object manipulation and task-relevance interacted in that location memory for relevant objects was superior to irrelevant ones only in the Handle condition. Including previous object recall performance as a covariate in the linear mixed-model analysis of times to first fixation allowed us to explore the interaction between remembered/forgotten object identities and the execution of location memory. Identity memory performance predicted location memory in the Find but not the Handle condition, suggesting that active object handling leads to strong spatial representations independent of object identity memory. We argue that object handling facilitates the prioritization of relevant location information, but this might come at the cost of deprioritizing irrelevant information.

Compiling irrelevant objects to counters. Special case of creation planning

Compiling irrelevant objects to counters. Special case of creation planning

Effect of tDCS on task relevant and irrelevant perceptual learning of complex objects.

During perceptual learning the visual representations in the brain are altered, but these changes' causal role has not yet been fully characterized. We used transcranial direct current stimulation (tDCS) to investigate the role of higher visual regions in lateral occipital cortex (LO) in perceptual learning with complex objects. We also investigated whether object learning is dependent on the relevance of the objects for the learning task. Participants were trained in two tasks: object recognition using a backward masking paradigm and an orientation judgment task. During both tasks, an object with a red line on top of it were presented in each trial. The crucial difference between both tasks was the relevance of the object: the object was relevant for the object recognition task, but not for the orientation judgment task. During training, half of the participants received anodal tDCS stimulation targeted at the lateral occipital cortex (LO). Afterwards, participants were tested on how well they recognized the trained objects, the irrelevant objects presented during the orientation judgment task and a set of completely new objects. Participants stimulated with tDCS during training showed larger improvements of performance compared to participants in the sham condition. No learning effect was found for the objects presented during the orientation judgment task. To conclude, this study suggests a causal role of LO in relevant object learning, but given the rather low spatial resolution of tDCS, more research on the specificity of this effect is needed. Further, mere exposure is not sufficient to train object recognition in our paradigm.

When Affect Supports Cognitive Control – A Working Memory Perspective

Abstract The paper delineates a study of executive functions (EFs), construed as procedural working memory (WM), from a motivational perspective. Since WM theories and motivation theories are both concerned with purposive activity, the role of implicit evaluations (affects) observed in goal pursuit can be anticipated to arise also in the context of cognitive control, e.g., during the performance of the Stroop task. The role of positive and negative affect in goal pursuit consists in controlling attention resources according to the goal and situational requirements. Positive affect serves to maintain goals and means in the scope of attention (EF1), whereas negative affect activates the inhibition of non-functional contents, e.g., distractors and irrelevant objects (resulting in attention disengagement; EF2). Adaptation to conflict proceeds via sequential triggering of negative and positive affect (EF3). Moreover, it was demonstrated that the focus on action or reflection changes the scope of contents subjected to implicit (affective) control. Therefore, I suggest that the motivational system, to a large extent, plays the role of the Central Executive. The paper opens a discussion and proposes studies on affective mechanisms of cognitive control.

The Automaticity of Affordance of Dangerous Object.

Objects observation automatically elicits the activation of a reach-to-grasp response specifically directed to interact with the object, which is termed affordance. Murphy, van Velzen, and de Fockert (2012) found that only when an irrelevant object receives sufficient attention, it can potentiate an action. However, it remains unclear whether the dangerous object would afford an action when it receives insufficient attention. In this study, we manipulated the perceptual load in a letter identification task. Participants were required to identify a target letter with the right or left hand while ignoring a neutral or dangerous graspable object. The target letter was presented either on its own (low perceptual load), alongside five non-target letters (high load), or alongside eight non-target letters (super high load). Under the low perceptual load, for both neutral and dangerous object, responses were faster when the action afforded by the ignored object was congruent (vs. incongruent) with the current target response (t(27) = 4.44, p < .001; t(27) = 7.99, p < .001, respectively). However, during the high perceptual load, for dangerous object, responses were slower when the action afforded by the ignored object was congruent (vs. incongruent) with the current target response (t(27) = 4.97, p < .001). There was not any effect for both neutral object and dangerous object under super high perceptual load. These results suggest the affordance of dangerous object is also sensitive to the perceptual load. An irrelevant dangerous object can't potentiate an action if it receives insufficient attention.

3D BUILDING FAÇADE RECONSTRUCTION USING HANDHELD LASER SCANNING DATA

Abstract. 3D The three dimensional building modelling has been an interesting topic of research for decades and it seems that photogrammetry methods provide the only economic means to acquire truly 3D city data. According to the enormous developments of 3D building reconstruction with several applications such as navigation system, location based services and urban planning, the need to consider the semantic features (such as windows and doors) becomes more essential than ever, and therefore, a 3D model of buildings as block is not any more sufficient. To reconstruct the façade elements completely, we employed the high density point cloud data that obtained from the handheld laser scanner. The advantage of the handheld laser scanner with capability of direct acquisition of very dense 3D point clouds is that there is no need to derive three dimensional data from multi images using structure from motion techniques. This paper presents a grammar-based algorithm for façade reconstruction using handheld laser scanner data. The proposed method is a combination of bottom-up (data driven) and top-down (model driven) methods in which, at first the façade basic elements are extracted in a bottom-up way and then they are served as pre-knowledge for further processing to complete models especially in occluded and incomplete areas. The first step of data driven modelling is using the conditional RANSAC (RANdom SAmple Consensus) algorithm to detect façade plane in point cloud data and remove noisy objects like trees, pedestrians, traffic signs and poles. Then, the façade planes are divided into three depth layers to detect protrusion, indentation and wall points using density histogram. Due to an inappropriate reflection of laser beams from glasses, the windows appear like holes in point cloud data and therefore, can be distinguished and extracted easily from point cloud comparing to the other façade elements. Next step, is rasterizing the indentation layer that holds the windows and doors information. After rasterization process, the morphological operators are applied in order to remove small irrelevant objects. Next, the horizontal splitting lines are employed to determine floors and vertical splitting lines are employed to detect walls, windows, and doors. The windows, doors and walls elements which are named as terminals are clustered during classification process. Each terminal contains a special property as width. Among terminals, windows and doors are named the geometry tiles in definition of the vocabularies of grammar rules. Higher order structures that inferred by grouping the tiles resulted in the production rules. The rules with three dimensional modelled façade elements constitute formal grammar that is named façade grammar. This grammar holds all the information that is necessary to reconstruct façades in the style of the given building. Thus, it can be used to improve and complete façade reconstruction in areas with no or limited sensor data. Finally, a 3D reconstructed façade model is generated that the accuracy of its geometry size and geometry position depends on the density of the raw point cloud.

Templates for rejection can specify semantic properties of nontargets in natural scenes.

In contrast to standard search templates that specify a target object's expected features, templates for rejection (TFR) may specify features of nontargets, biasing attention away from irrelevant objects. Little is known about TFR, and virtually nothing is known about their role in guiding search across natural scenes. In such scenes, targets and nontargets may not be easily distinguished on the basis of their visual features; it has been claimed that standard search templates may therefore specify target objects' semantic features to guide attention. Here, we ask whether TFR can do so. Noting a limitation of previous procedures used to study standard search templates, we trialed an alternative method to examine semantic templates for nontarget exclusion in natural scene search. We found that when nontargets belonged unpredictably to either of two physically distinct categories, search was less efficient than when targets belonged to one known category. This two-category cost, attributed to inefficient application of search templates, was absent for two physically dissimilar but semantically related categories. Adding a training phase to highlight semantic distinctiveness of two object categories reinstated the two-category cost, precluding stimulus-based accounts of the effect. These patterns were not observed for one-image displays or when observers searched for object categories rather than ignoring them, demonstrating their specificity to TFR, the inadequacy of search-and-destroy models to account for them, and likely basis in attentional guidance. TFR can specify semantic information to guide attention away from nontargets.

Impact of Wet Macular Degeneration on the Execution of Natural Actions.

To use eye movements to investigate how people with a central scotoma might be impaired in the execution of natural actions and whether task familiarity affects performance. Sixteen participants with AMD and 16 age-matched controls performed two natural actions: (1) a familiar sandwich-making task and (2) a less familiar model-building task. In each action, task-relevant and task-irrelevant objects were placed on a table, covering 90°. The participants were asked to execute the actions without a time constraint. Eye movements were recorded. The people with AMD were significantly slower than the controls, both in the exploration phase (before the first reaching movement) and in the working phase (execution of action), especially in the unfamiliar task. Gaze duration was longer on relevant than irrelevant objects in both groups and tasks, as might be expected. However, for the participants with AMD, gaze durations were longer on all of the objects, whether relevant or irrelevant, except in the more familiar task. This suggests that participants with AMD take longer to extract the information they need but that this can be counteracted when the task items are familiar. The number of saccades/min of the task was significantly greater for the people with AMD than for the controls. The present results show that people with AMD can accomplish natural actions efficiently, but need longer gaze durations and more eye movements than normally sighted people. This effect can be reduced when executing a familiar task.

Stimulus-response correspondence in go-nogo and choice tasks: Are reactions altered by the presence of an irrelevant salient object?

In 2-choice tasks, responses are faster when stimulus location corresponds to response location, even when stimulus location is irrelevant. Dolk et al. (J Exp Psychol Hum Percept Perform 39:1248-1260, 2013a) found this stimulus-response correspondence effect with a single response location in a go-nogo task when an irrelevant Japanese waving cat was present. They argued that salient objects trigger spatial coding of the response relative to that object. We examined this claim using both behavioral and lateralized readiness potential (LRP) measures. In Experiment 1 participants determined the pitch of a left- or right-positioned tone, whereas in Experiment 2 they determined the color of a dot within a centrally located hand pointing left, right, or straight ahead. In both experiments, participants performed a go-nogo task with the right-index finger and a 2-choice task with both index fingers, with a left-positioned Japanese waving cat present or absent. For the go-nogo task, the cat induced a correspondence effect on response times (RT) to the tones (Experiment 1) but not the visual stimuli (Experiment 2). For the 2-choice task, a correspondence effect was evident in all conditions in both experiments. Cat's presence/absence did not significantly modulate the effect for right and left responses, although there was a trend toward increased RT and LRP for right responses in Experiment 1. The results imply that a salient, irrelevant object could provide a reference frame for response coding when attention is available to process it, as is likely in an auditory task (Experiment 1) but not a visual task (Experiment 2).

Probing the Cognitive Mechanism of Mental Representational Change During Chunk Decomposition: A Parametric fMRI Study.

Chunk decomposition plays an important role in cognitive flexibility in particular with regards to representational change, which is critical for insight problem solving and creative thinking. In this study, we investigated the cognitive mechanism of decomposing Chinese character chunks through a parametric fMRI design. Our results from this parametric manipulation revealed widely distributed activations in frontal, parietal, and occipital cortex and negative activations in parietal and visual areas in response to chunk tightness during decomposition. To mentally manipulate the element of a given old chunk, superior parietal lobe appears to support element restructuring in a goal-directed way, whereas the negatively activated inferior parietal lobe may support preventing irrelevant objects from being attended. Moreover, determining alternative ways of restructuring requires a constellation of frontal areas in the cognitive control network, such as the right lateral prefrontal cortex in inhibiting the predominant chunk representations, the presupplementary motor area in initiating a transition of mental task set, and the inferior frontal junction in establishing task sets. In conclusion, this suggests that chunk decomposition reflects mental transformation of problem representation from an inappropriate state to a new one alongside with an evaluation of novel and insightful solutions by the caudate in the dorsal striatum.

Functional fixedness: The functional significance of delayed disengagement based on attention set.

During search, the disengagement of attention is automatically delayed when a fixated but task-irrelevant object shares features of the search target. We examined whether delayed disengagement based on top-down attention set is potentially functional, resulting in additional processing of the fixated item. To accomplish this, we adapted the oculomotor disengagement paradigm. Participants saccaded to a peripheral object of a particular color and responded to the identity of the letter within it. To initiate search participants made a saccade away from an always irrelevant object at the center of the screen that matched or mismatched the target's color and contained a letter that was congruent or incongruent with the target letter. We found that delayed disengagement based on attention set was associated with deeper processing of the center item: a congruency effect between the center letter and peripheral target letter was only observed when the center object's color matched participants' attention set. Results are consistent with the proposal that delayed disengagement based on attention set is functionally significant, automatically encouraging deeper levels of processing of target-like objects that fall within the focus of attention.

Autogrouped Sparse Representation for Visual Analysis

In image classification, recognition or retrieval systems, image contents are commonly described by global features. However, the global features generally contain noise from the background, occlusion, or irrelevant objects in the images. Thus, only part of the global feature elements is informative for describing the objects of interest and useful for the image analysis tasks. In this paper, we propose algorithms to automatically discover the subgroups of highly correlated feature elements within predefined global features. To this end, we first propose a novel mixture sparse regression (MSR) method, which groups the elements of a single vector according to the membership conveyed by their sparse regression coefficients. Based on MSR, we proceed to develop the autogrouped sparse representation (ASR), which groups correlated feature elements together through fusing their individual sparse representations over multiple samples. We apply ASR/MSR in two practical visual analysis tasks: 1) multilabel image classification and 2) motion segmentation. Comprehensive experimental evaluations show that our proposed methods are able to achieve superior performance compared with the state-of-the-art classification on these two tasks.

Automatic crack detection and classification method for subway tunnel safety monitoring.

Cracks are an important indicator reflecting the safety status of infrastructures. This paper presents an automatic crack detection and classification methodology for subway tunnel safety monitoring. With the application of high-speed complementary metal-oxide-semiconductor (CMOS) industrial cameras, the tunnel surface can be captured and stored in digital images. In a next step, the local dark regions with potential crack defects are segmented from the original gray-scale images by utilizing morphological image processing techniques and thresholding operations. In the feature extraction process, we present a distance histogram based shape descriptor that effectively describes the spatial shape difference between cracks and other irrelevant objects. Along with other features, the classification results successfully remove over 90% misidentified objects. Also, compared with the original gray-scale images, over 90% of the crack length is preserved in the last output binary images. The proposed approach was tested on the safety monitoring for Beijing Subway Line 1. The experimental results revealed the rules of parameter settings and also proved that the proposed approach is effective and efficient for automatic crack detection and classification.