Specific Target Objects Research Articles

Visual perceptual learning is enhanced by training in the illusory far space.

Visual objects in the peripersonal space (PPS), are perceived faster than farther ones, appearing in the extrapersonal space (EPS). This shows preferential processing for visual stimuli near our body. Such an advantage should favor visual perceptual learning occurring near, as compared to far from observers, but opposite evidence has been recently provided from online testing protocols, showing larger perceptual learning in the far space. Here, we ran two laboratory-based experiments investigating whether visual training in PPS and EPS has different effects. We used the horizontal Ponzo Illusion to create a lateralized depth perspective while participants completed a visual search task in which they reported whether or not a specific target object orientation (e.g., a triangle pointing upward) was present amongst distractors. This task was completed before and after a training phase in either the (illusory) near or far space for one hour. In Experiment 1, the near space was in the left hemispace, whereas in Experiment 2 it was in the right. Results showed that, in both experiments, participants were more accurate after training in the far space, whereas training in the near space led to either improvement in the far space (Exp. 1), or no change (Exp. 2). Moreover, we found a larger visual perceptual learning when stimuli were presented in the left compared to the right hemispace. Differently from visual processing, visual perceptual learning is more effective in the far space. We propose that depth is a key dimension that can be used to improve human visual learning.

Partitioning intensity inhomogeneity colour images via Saliency-based active contour

Partitioning or segmenting intensity inhomogeneity colour images is a challenging problem in computer vision and image shape analysis. Given an input image, the active contour model (ACM) which is formulated in variational framework is regularly used to partition objects in the image. A selective type of variational ACM approach is better than a global approach for segmenting specific target objects, which is useful for applications such as tumor segmentation or tissue classification in medical imaging. However, the existing selective ACMs yield unsatisfactory outcomes when performing the segmentation for colour (vector-valued) with intensity variations. Therefore, our new approach incorporates both local image fitting and saliency maps into a new variational selective ACM to tackle the problem. The euler-lagrange (EL) equations were presented to solve the proposed model. Thirty combinations of synthetic and medical images were tested. The visual observation and quantitative results show that the proposed model outshines the other existing models by average, with the accuracy of 2.23% more than the compared model and the Dice and Jaccard coefficients which were around 12.78% and 19.53% higher, respectively, than the compared model.

Visual perceptual learning is effective in the illusory far but not in the near space

Visual shape discrimination is faster for objects close to the body, in the peripersonal space (PPS), compared with objects far from the body. Visual processing enhancement in PPS occurs also when perceived depth is based on 2D pictorial cues. This advantage has been observed from relatively low-level (detection, size, orientation) to high-level visual features (face processing). While multisensory association also displays proximal advantages, whether PPS influences visual perceptual learning remains unclear. Here, we investigated whether perceptual learning effects vary according to the distance of visual stimuli (near or far) from the observer, illusorily induced by leveraging the Ponzo illusion. Participants performed a visual search task in which they reported whether a specific target object orientation (e.g., triangle pointing downward) was present among distractors. Performance was assessed before and after practicing the visual search task (30 minutes/day for 5 days) at either the close (near group) or far (far group) distance. Results showed that participants that performed the training in the near space did not improve. By contrast, participants that performed the training in the far space showed an improvement in the visual search task in both the far and near spaces. We suggest that such improvement following the far training is due to a greater deployment of attention in the far space, which could make the learning more effective and generalize across spaces.

Improvement in Error Recognition of Real-Time Football Images by an Object-Augmented AI Model for Similar Objects

In this paper, we analyze the recognition error of the general AI recognition model and propose the structure-modified and object-augmented AI recognition model. This model has object detection features that distinguish specific target objects in target areas where players with similar shapes and characteristics overlapped in real-time football images. We implemented the AI recognition model by reinforcing the training dataset and augmenting the object class. In addition, it is shown that the recognition rate increased by modifying the model structure based on the recognition errors analysis of general AI recognition models. Recognition errors were decreased by applying the modules of HSV processing and differentiated classes (overlapped player groups) learning to the general AI recognition model. We experimented in order to compare the recognition error reducing performance with the general AI model and the proposed AI model by the same real-time football images. Through this study, we have confirmed that the proposed model can reduce errors. It showed that the proposed AI model structure to recognize similar objects in real-time and in various environments could be used to analyze football games.

Using Transfer Learning to Reduce Preparing Time of Online Diagnosis with Limited Rotating-Component Data

Under the environment of mass-customization, valid samples for a specific target object become fewer and fewer. Thus, Transfer Learning (TL) is widely applied to those data-consuming Deep Learning (DL) and Machine Learning (ML) models to avoid recollecting high-volume training samples from scratch. However, TL-based methods used in fault diagnosis for mechanical rotating components are still in the infancy phase. Poor practicality and usability problems increase the total preparing time. Without online diagnosis, there might be more unknown risks during the preparing time of a TL method. To reduce the preparing time of existing TL techniques, this paper proposes a TL-based Gradual Improvement Scheme (TLGIS) to achieve a systematic online diagnosis of rotating components. TLGIS solves the problem through seamless interactions among five modules of Data Preprocessing (DP), Data Estimating (DE), Data Modeling (DM), Model Improving (MI), and Accuracy Checking, (AC). Via illustrative examples, TLGIS is demonstrated to be reliable and robust using deep Convolutional Neural Networks (CNN) by two public bearing datasets from the Internet and two types of rotating wheel datasets from a practical CNC machine. In total, TLGIS only requires at least 80 samples and at most 250 samples divided into small-size batches to gradually and continuously fine-tune the model with an accuracy of 90% and 99%, respectively, instead of inputting a large-scale dataset all at once, which is unrealistic in a practical production scenario. Compared to conventional approaches, TLGIS efficiently saves at most 90% of preparing time.

Sovereign anxiety in Myanmar: An emotional geopolitics of China's Belt and Road Initiative

AbstractGeographers have increasingly attended to the role of emotion in geopolitical encounters and the geopolitics of cross‐border infrastructure projects. While scholars have theorised fear as an emotion produced by elite geopolitical discourses and encounters between bodies, we know much less about how infrastructure's materialities provoke fear and anxiety. Furthermore, key distinctions between anxiety – or a psychological state of insecurity and unease – and fear, which is attached to a specific target object, are still not fully understood. Focused on the uncertainties over China's Belt and Road Initiative (BRI), we develop the concept of sovereign anxiety – a generalised condition of unease over the security of one's political community – to account for how the BRI generates not only the hard materials of infrastructure (e.g., roads, dams and pipelines), but also the social practices of affect and emotion. Sovereign anxiety, we argue, is heightened by the absence of transparency over China's infrastructure investments in Myanmar. In this paper, we trace how sovereign anxiety is variously experienced and grounded in residents' observations, personal biographies, social histories, and sense of community belonging. We also identify three themes by which fears of the BRI are articulated: relations, roads, and resources. This article contributes an emotional geopolitics perspective to grounded studies of the BRI, while also demonstrating the geopolitical significance of attending to the emotional lives of infrastructure in relation to the BRI and beyond.

Developing a computer vision system for real-time color measurement – A case study with color characterization of roasted rice

To take advantage of the consistent and automated measurement capability of Computer Vision Systems (CVS), an initial CVS with inclined imaging of the food object was developed to measure its color in a real-time food process. The RGB-L*a*b* color-space transformation model of the CVS was obtained by multiple linear regression. The impact of training data and regularization for high-order regression models was thoroughly evaluated using cross-validation results and mean color error. A case study with roasted rice highlighted the importance of the training data size, the choice of samples in the appropriate color gamut, and the inclusion of representative color samples for a specific target object. The optimal cubic Ridge regression model had the CIEDE2000 mean color errors of 1.73, 1.11, and 1.33 when respectively evaluated with the training dataset and two testing rice datasets. The higher testing accuracy reconfirmed the appropriate training data preparation and demonstrated the benefits of regularization in regression. Based on various performance indices, the adopted model outperformed those reported in the literature. Therefore, the proposed CVS could be reproduced for color characterization of food products in automated and real-time processes. • A CVS was developed for food color measurement in potential real-time food processes that required inclined imaging. • More training data in the appropriate color gamut with representatives of the target object could improve the model accuracy. • A cubic Ridge regression model was optimally obtained for accurate and reliable color-space transformation. • Comparably small CIEDE2000 mean color errors of 1.11 and 1.33 were obtained with two testing rice datasets. • Research results could be reproduced for color characterization of food products in automated and real-time processes.

Efficient compressive holographic reconstruction based on hologram segmentation.

Compressive holography can successfully reconstruct a three-dimensional layered object from a two-dimensional hologram. However, the extremely time-consuming reconstruction limits its range of applications. We propose a dimension reduction of measurement matrix (DRMM) method to accelerate compressive holographic reconstruction. The calculation time is substantially reduced while the reconstruction quality is improved by DRMM, which is implemented by a hologram segmentation approach and a parallel computing technique. Holograms of specific target objects are segmented from the hologram of a three-dimensional layered object, and the reconstruction can be implemented in parallel using multicore processors. We present both simulation and experimental results to show the feasibility and effectiveness of the proposed method.

Side-Scan Sonar Image Classification Based on Style Transfer and Pre-Trained Convolutional Neural Networks

Side-scan sonar is widely used in underwater rescue and the detection of undersea targets, such as shipwrecks, aircraft crashes, etc. Automatic object classification plays an important role in the rescue process to reduce the workload of staff and subjective errors caused by visual fatigue. However, the application of automatic object classification in side-scan sonar images is still lacking, which is due to a lack of datasets and the small number of image samples containing specific target objects. Secondly, the real data of side-scan sonar images are unbalanced. Therefore, a side-scan sonar image classification method based on synthetic data and transfer learning is proposed in this paper. In this method, optical images are used as inputs and the style transfer network is employed to simulate the side-scan sonar image to generate “simulated side-scan sonar images”; meanwhile, a convolutional neural network pre-trained on ImageNet is introduced for classification. In this paper, we experimentally demonstrate that the maximum accuracy of target classification is up to 97.32% by fine-tuning the pre-trained convolutional neural network using a training set incorporating “simulated side-scan sonar images”. The results show that the classification accuracy can be effectively improved by combining a pre-trained convolutional neural network and “similar side-scan sonar images”.

Local reliability weighting explains identification of partially masked objects in natural images

A fundamental natural visual task is the identification of specific target objects in the environments that surround us. It has long been known that some properties of the background have strong effects on target visibility. The most well-known properties are the luminance, contrast, and similarity of the background to the target. In previous studies, we found that these properties have highly lawful effects on detection in natural backgrounds. However, there is another important factor affecting detection in natural backgrounds that has received little or no attention in the masking literature, which has been concerned with detection in simpler backgrounds. Namely, in natural backgrounds the properties of the background often vary under the target, and hence some parts of the target are masked more than others. We began studying this factor, which we call the “partial masking factor,” by measuring detection thresholds in backgrounds of contrast-modulated white noise that was constructed so that the standard template-matching (TM) observer performs equally well whether or not the noise contrast modulates in the target region. If noise contrast is uniform in the target region, then this TM observer is the Bayesian optimal observer. However, when the noise contrast modulates then the Bayesian optimal observer weights the template at each pixel location by the estimated reliability at that location. We find that human performance for modulated noise backgrounds is predicted by this reliability-weighted TM (RTM) observer. More surprisingly, we find that human performance for natural backgrounds is also predicted by the RTM observer.

Pixel super-resolution lensless in-line holographic microscope with hologram segmentation

We propose a resolution enhancement method for a lensless in-line holographic microscope (LIHM) by combining the hologram segmentation and pixel super-resolution (PSR) techniques. Our method is suitable for imaging specific target objects in samples, where the in-line hologram is disturbed by other objects in the samples. The resolution-enhancement capability of our method was proved by numerical simulations and imaging experiments while using a standard resolution target in a two-layer setup. We also applied our LIHM system to image the sample of living algae Euglena gracilis in water solution for further demonstration.

A Region Merging Segmentation with Local Scale Parameters: Applications to Spectral and Elevation Data

Region merging is the most effective method for the segmentation of remote sensing data. The quality and the size of the resulted image objects is controlled by a global heterogeneity threshold, termed as the scale parameter. However, the multidimensional nature of the visible features in a scene defies the use of an even optimum single global scale parameter. In this study, a novel region merging segmentation method is proposed, where a local scale parameter is defined for each image object by its internal and external heterogeneity measures (i.e., local variance and Moran’s I). This method allows image objects with low internal and external heterogeneity to be further merged with higher scale parameter values, since they are more likely to be a part of an adjacent object, than objects with high internal and external heterogeneity. The proposed method was applied in spectral and elevation data and its results were evaluated visually and with supervised and unsupervised evaluation methods. The comparison with multi-resolution segmentation (MRS) showed that the proposed region merging method can produce improved segmentation results in terms of maximizing intra-object homogeneity and inter-object heterogeneity as well as in the delimitation of specific target objects, present in spectral and elevation data. The unsupervised evaluation results of the (1) Côte d’Azur, (2) Manchester, and (3) Szada images from the SZTAKI-INRIA building detection dataset showed that the proposed method (overall goodness, OGf (1): 0.7375, (2): 0.7923, (3): 0.7967) performs better than MRS (OGf (1): 0.7224, (2): 0.7648, (3): 0.7823). The higher values of OGf indicate their ability to produce segmentation results with reduced over-segmentation effects and without the need of presegmented input data, in contrast to the objective heterogeneity and relative homogeneity (OHRH) hybrid segmentation method (OGf (1): 0.5864, (2): 0.5151, (3): 0.6983).

PERCEPTUAL GUIDANCE IMPROVES VISUAL SEARCH IN AGING AND IN ALZHEIMER’S DISEASE: EVIDENCE FROM EYE-MOVEMENT BEHAVIOR WITHIN REAL-WORLD SCENES

Elderly people and, especially, patients with Alzheimer's disease (AD) show impaired performance in visual search. This seems to arise from inefficient strategies, with an increased number of eye fixations and longer fixation durations. As many activities of daily living require efficient search of objects within complex environments, visual search deficits of AD patients may have a key role in reducing their functional autonomy. However, while previous studies have examined patients’ search in simple object arrays, little is know about their search behaviour in real-world scenes. Using an EyeLink 1000, we recorded eye movements of 12 mild AD patients, 24 healthy age-matched individuals (“elderly controls”) and 32 young individuals (“young controls”) during search within real-world scenes. We examined whether any group difference in eye guidance during search depends on target template specificity (name vs. precise picture), target visual salience (low vs. high) and reliability of scene semantics (consistent vs. inconsistent targets). We analysed data using linear mixed effect models, considering search initiation (first search saccade), scene scanning (until first target fixations) and target verification (dwell time on target). Aging (elderly vs. young controls) and AD (patients vs. elderly controls) led to less effective behaviour in all search phases, with fewer first saccades directed toward the target, longer scanning time and longer verification time. While these patterns were not affected by scene semantic guidance, a greater amount of target perceptual guidance was especially beneficial for the elderly controls and, even more, for the AD patients. In particular, higher target salience reduced performance gaps between elderly and young controls during scene scanning and target verification, and – when combined with a pictorial template – between patients and elderly controls during scene scanning. Eye-movement analysis enables to examine the time course of fine impairments during real-world scene search. Enhancing target visual salience and providing a priori knowledge of the perceptual appearance of the specific target objects seem promising ways of improving visual search behaviour in healthy elderly individuals and in AD patients. This understanding may contribute to the development of more appropriate information systems and living places, promoting individual's everyday functional autonomy.

Situated reference resolution using visual saliency and crowdsourcing-based priors for a spoken dialog system within vehicles

In this paper, we address issues in situated language understanding in a moving car. More specifically, we propose a reference resolution method to identify user queries about specific target objects in their surroundings. We investigate methods of predicting which target object is likely to be queried given a visual scene and what kind of linguistic cues users naturally provide to describe a given target object in a situated environment. We propose methods to incorporate the visual saliency of the visual scene as a prior. Crowdsourced statistics of how people describe an object are also used as a prior. We have collected situated utterances from drivers using our research system, which was embedded in a real vehicle. We demonstrate that the proposed algorithms improve target identification rate by 15.1% absolute over the baseline method that does not use visual saliency-based prior and depends on public database with a limited number of category information.

Person Re-identification using Sparse Representation with a Saliency-weighted Dictionary

Intelligent video surveillance systems have been developed to monitor global areas and find specific target objects using a large-scale database. However, person re-identification presents some challenges, such as pose change and occlusions. To solve the problems, this paper presents an improved person re-identification method using sparse representation and saliency-based dictionary construction. The proposed method consists of three parts: i) feature description based on salient colors and textures for dictionary elements, ii) orthogonal atom selection using cosine similarity to deal with pose and viewpoint change, and iii) measurement of reconstruction error to rank the gallery corresponding a probe object. The proposed method provides good performance, since robust descriptors used as a dictionary atom are generated by weighting some salient features, and dictionary atoms are selected by reducing excessive redundancy causing low accuracy. Therefore, the proposed method can be applied in a large scale–database surveillance system to search for a specific object.

Video tracking using learned hierarchical features.

In this paper, we propose an approach to learn hierarchical features for visual object tracking. First, we offline learn features robust to diverse motion patterns from auxiliary video sequences. The hierarchical features are learned via a two-layer convolutional neural network. Embedding the temporal slowness constraint in the stacked architecture makes the learned features robust to complicated motion transformations, which is important for visual object tracking. Then, given a target video sequence, we propose a domain adaptation module to online adapt the pre-learned features according to the specific target object. The adaptation is conducted in both layers of the deep feature learning module so as to include appearance information of the specific target object. As a result, the learned hierarchical features can be robust to both complicated motion transformations and appearance changes of target objects. We integrate our feature learning algorithm into three tracking methods. Experimental results demonstrate that significant improvement can be achieved using our learned hierarchical features, especially on video sequences with complicated motion transformations.

Item and category-based attentional control during search for real-world objects: Can you find the pants among the pans?

To compare the speed and efficiency of item-based and category-based attentional control during visual search for real-world objects, we measured N2pc components as electrophysiological markers of attentional target selection. In different blocks, participants searched for 1 or 2 specific target objects or for any object in a target category (items of clothing or kitchen objects). Search displays contained 6 line drawings of different objects, and targets always appeared together with 5 distractors from the other object category. The presence of N2pc components to categorically defined targets demonstrated that category-based search can operate at visuoperceptual processing stages. In contrast to previous findings for letter/digit search (Nako, Wu, & Eimer, 2014), target N2pc components were delayed by 40 ms during category-guided search relative to single-target search. This suggests that for objects and object categories that are less familiar than alphanumerical stimuli, category-guided target selection operates less efficiently than selection that is based on a physical match with an attentional template. (PsycINFO Database Record (c) 2014 APA, all rights reserved).

Navigation and exploration of an urban virtual environment by children with autism spectrum disorder compared to children with typical development

Autism spectrum disorder (ASD) is a severe disorder therefore the importance to implement targeted interventions in order to improve daily life of children with ASD. For this purpose, virtual environments (VEs), i.e., simulations of the real world based on 3D computer graphics, can offer a safe learning environment for them. This study analyzed navigation and exploration of an urban VE by children with ASD in comparison to children with a typical development. Sixteen children with ASD and 16 matched control ones were involved. After an initial training phase, children carried out two tasks: the first one was navigating in an unfamiliar urban environment which they could freely explore; the second one was navigating in the same environment but with the goal of finding specific target objects, as in a treasure hunt. In the first task, children with ASD spent significantly less time in active exploration and explored fewer zones than controls. No differences were found between the two groups in the second task. Our data indicate that, when freely exploring an unfamiliar VE, children with ASD explore less the environment compared to the control ones. By repeating the exploration with a game-like goal, no differences were found instead. Neuropsychological and motivational aspects should be considered in order to explain these findings.

Real-Time Continuous Image Registration Enabling Ultraprecise 2-D Motion Tracking

In this paper, we present a novel continuous image registration method (CIRM), which yields near-zero bias and has high computational efficiency. It can be realized for real-time position estimation to enable ultraprecise 2-D motion tracking and motion control over a large motion range. As the two variables of the method are continuous in spatial domain, pixel-level image registration is unnecessary, thus the CIRM can continuously track the moving target according to the incoming target image. When applied to a specific target object, measurement resolution of the method is predicted according to the reference image model of the object along with the variance of the camera's overall image noise. The maximum permissible target speed is proportional to the permissible frame rate, which is limited by the required computational time. The precision, measurement resolution, and computational efficiency of the method are verified through computer simulations and experiments. Specifically, the CIRM is implemented and integrated with a visual sensing system. Near-zero bias, measurement resolution of 0.1 nm (0.0008 pixels), and measurement of one nanometer stepping are demonstrated.

An Experiment on Behavior Generalization and the Emergence of Linguistic Compositionality in Evolving Robots

Populations of simulated agents controlled by dynamical neural networks are trained by artificial evolution to access linguistic instructions and to execute them by indicating, touching, or moving specific target objects. During training the agent experiences only a subset of all object/action pairs. During postevaluation, some of the successful agents proved to be able to access and execute also linguistic instructions not experienced during training. This owes to the development of a semantic space, grounded in the sensory motor capability of the agent and organized in a systematized way in order to facilitate linguistic compositionality and behavioral generalization. Compositionality seems to be underpinned by a capability of the agents to access and execute the instructions by temporally decomposing their linguistic and behavioral aspects into their constituent parts (i.e., finding the target object and executing the required action). The comparison between two experimental conditions, in one of which the agents are required to ignore rather than to indicate objects, shows that the composition of the behavioral set significantly influences the development of compositional semantic structures.