Multiple Similar Objects Research Articles

AI-Based Intelligent Monitoring System for Estrus Prediction in the Livestock Industry

In order to improve a livestock breeding environment that considers securing safe cattle resources and improving productivity for the intelligent farm, we propose an animal-friendly and worker-friendly intellectual monitoring system with Artificial Intelligent (AI) technology. In order to secure safe cattle resources and increase productivity for the livestock industry, it is necessary to secure the self-activities of the cattle and predict the estrous state of target cattle as quickly as possible. For the prediction of the estrous state, it is necessary to continuously observe the cattle behavior by workers and quantify the behavior of the target cattle, but that is not easy for workers and needs a long period of continuous observation. We developed the intelligent monitoring system (IMS) with the ARM (Augmented Recognition Model) for the intelligent farm that can predict the estrus of target cattle and get activity data for individual cattle, and then the system was applied to a typical cattle farm for activity monitoring of the Korean cattle (Hanwoo). Therefore, we confirmed the target Hanwoo group with more than 400 activities among the Hanwoo groups using the ARM threshold. Thus, we verified the potential of the proposed system for tracking multiple similar objects.

Adaptive Selection of Reference Frames for Video Object Segmentation.

Video object segmentation is a challenging task in computer vision because the appearances of target objects might change drastically along the time in the video. To solve this problem, space-time memory (STM) networks are exploited to make use of the information from all the intermediate frames between the first frame and the current frame in the video. However, fully using the information from all the memory frames may make STM not practical for long videos. To overcome this issue, a novel method is developed in this paper to select the reference frames adaptively. First, an adaptive selection criterion is introduced to choose the reference frames with similar appearance and precise mask estimation, which can efficiently capture the rich information of the target object and overcome the challenges of appearance changes, occlusion, and model drift. Secondly, bi-matching (bi-scale and bi-direction) is conducted to obtain more robust correlations for objects of various scales and prevents multiple similar objects in the current frame from being mismatched with the same target object in the reference frame. Thirdly, a novel edge refinement technique is designed by using an edge detection network to obtain smooth edges from the outputs of edge confidence maps, where the edge confidence is quantized into ten sub-intervals to generate smooth edges step by step. Experimental results on the challenging benchmark datasets DAVIS-2016, DAVIS-2017, YouTube-VOS, and a Long-Video dataset have demonstrated the effectiveness of our proposed approach to video object segmentation.

An explicit investigation of the roles that feature distributions play in rapid visual categorization.

Ensemble representations are often described as efficient tools when summarizing features of multiple similar objects as a group. However, it can sometimes be more useful not to compute a single summary description for all of the objects if they are substantially different, for example when they belong to entirely different categories. It was proposed that the visual system can efficiently use the distributional information of ensembles to decide whether simultaneously displayed items belong to single or several different categories. Here we directly tested how the feature distribution of items in a visual array affects an ability to discriminate individual items (Experiment 1) and sets (Experiments 2-3) when participants were instructed explicitly to categorize individual objects based on the median of size distribution. We varied the width (narrow or fat) as well as the shape (smooth or two-peaked) of distributions in order to manipulate the ease of ensemble extraction from the items. We found that observers unintentionally relied on the grand mean as a natural categorical boundary and that their categorization accuracy increased as a function of the size differences among individual items and a function of their separation from the grand mean. For ensembles drawn from two-peaked size distributions, participants showed better categorization performance. They were more accurate at judging within-category ensemble properties in other dimensions (centroid and orientation) and less biased by superset statistics. This finding corroborates the idea that the two-peaked feature distributions support the "segmentability" of spatially intermixed sets of objects. Our results emphasize important roles of ensemble statistics (mean, range, distribution shape) in explicit visual categorization.

Mutation based self regulating and self perception particle swarm optimization for efficient object tracking in a video

Object tracking is a pre-eminent task in many computer vision application.This paper addresses a realistic object tracking problem, considering single as well as multiple objects with different challenging perspectives. In view of this, a Mutation-based Self Regulating and self-Perception Particle Swarm Optimization (MSRSP-PSO) algorithm is proposed in order to explore the effective sampling strategy as well as to estimate and localize the optimum position of the target object. In conjunction with the optimization technique, an effective appearance model is also used for the reinforcement of tracker to achieve better accuracy. The appearance model is constructed with the joint distribution of the Local Binary Pattern (LBP)/Local Contrast Measure (LCM) based texture description and the Ohta color features. In pursuance of validating the efficacy of the proposed tracking algorithm, performance of other state-of-the-art methods such as, Particle Swarm Optimization (PSO) based tracking, Sequential Particle Swarm Optimization (SPSO) based tracking, and Adaptive Gaussian Particle Swarm Optimization (AGPSO) based tracking algorithms are compared with the proposed tracking algorithm. Both the quantitative and qualitative evaluation of the experiments, validate the effectiveness of the proposed technique.The performance measures include the convergence rate, center location error and the coverage test (PASCAL SCORE). Furthermore, the non-parametric statistical analysis using the Sign test and Wilcoxon signed rank test, is also figured out to validate the statistical superiority of the proposed MSRSP-PSO algorithm over others. The experimental results reveal that the proposed algorithm robustly track single as well as multiple similar objects with complex interaction.

Rank-1 Tensor Approximation for High-Order Association in Multi-target Tracking

High-order motion information is important in multi-target tracking (MTT) especially when dealing with large inter-target ambiguities. Such high-order information can be naturally modeled as a multi-dimensional assignment (MDA) problem, whose global solution is however intractable in general. In this paper, we propose a novel framework to the problem by reshaping MTT as a rank-1 tensor approximation problem (R1TA). We first show that MDA and R1TA share the same objective function and similar constraints. This discovery opens a door to use high-order tensor analysis for MTT and suggests the exploration of R1TA. In particular, we develop a tensor power iteration algorithm to effectively capture high-order motion information as well as appearance variation. The proposed algorithm is evaluated on a diverse set of datasets including aerial video sequences containing ariel borne dense highway scenes, top-view pedestrian trajectories, multiple similar objects, normal view pedestrians and vehicles. The effectiveness of the proposed algorithm is clearly demonstrated in these experiments.

Multi-Attributes Web Objects Classification based on Class-Attribute Relation Patterns Learning Approach

The amount of Web data increases with the proliferation of a variety of Web objects, primarily in the form of text, images, video, and music data files. Each of these published objects has some properties that support defining its class properties. Because of their diversity, using these attributes to learn and generate patterns for precise classification is very complicated. Even learning a set of attributes that clearly categorize the categories is very important. Existing attribute learning methods only learn attributes that are closely related to multiple similar objects, but if similar class objects have different attributes, this problem is difficult to learn and classify them. In this paper, a Multi-attributes Web Objects Classification (MA-WOC) based on Class-attribute Relation Patterns Learning Approach is being proposed, which generates a class-attribute with its multi relations patterns. The MA-WOC calculates the relationship probabilities of the attributes and the associated values of the class to understand the degree of association of the construction of classification pattern. To evaluate the effectiveness of the classifier, this will compare to an existing classifier that supports a multi-attribute data set, which shows improvisation of precision with a significant minimum Hamming loss. To evaluate the effectiveness of MA-WOC classification a comparison among the classifiers that are supported to the multi-attribute dataset are being performed to measure the accuracy and hamming loss.

New models and algorithms for the container stack rearrangement problem by yard cranes in maritime ports

Whenever sets of objects are piled up in heaps, columns or stacks, any rearrangement of these objects requires substantial amount of time, effort and cost. In this paper, we study the problem of optimizing such “stack-rearrangement operations” when multiple similar objects or blocks need to be rearranged. We discuss the problem in the context of yard crane operations in maritime ports, where cargo containers are stored and fetched by yard cranes. We intend to minimize the crane operations that are required during container stack rearrangement by yard cranes. This paper defines the underlying abstract mathematical problem and proves its computational complexity. Thereafter, we propose suitable mathematical models for the problem and devise several (exact) approaches to solve it. We provide several families of new problem data instances and the prove the efficacy of our algorithms by doing extensive computational analysis over the data instances.

Measuring and Predicting Visual Importance of Similar Objects.

Similar objects are ubiquitous and abundant in both natural and artificial scenes. Determining the visual importance of several similar objects in a complex photograph is a challenge for image understanding algorithms. This study aims to define the importance of similar objects in an image and to develop a method that can select the most important instances for an input image from multiple similar objects. This task is challenging because multiple objects must be compared without adequate semantic information. This challenge is addressed by building an image database and designing an interactive system to measure object importance from human observers. This ground truth is used to define a range of features related to the visual importance of similar objects. Then, these features are used in learning-to-rank and random forest to rank similar objects in an image. Importance predictions were validated on 5,922 objects. The most important objects can be identified automatically. The factors related to composition (e.g., size, location, and overlap) are particularly informative, although clarity and color contrast are also important. We demonstrate the usefulness of similar object importance on various applications, including image retargeting, image compression, image re-attentionizing, image admixture, and manipulation of blindness images.

Prostate contours delineation using interactive directional active contours model and parametric shape prior model.

Prostate contours delineation on Magnetic Resonance (MR) images is a challenging and important task in medical imaging with applications of guiding biopsy, surgery and therapy. While a fully automated method is highly desired for this application, it can be a very difficult task due to the structure and surrounding tissues of the prostate gland. Traditional active contours-based delineation algorithms are typically quite successful for piecewise constant images. Nevertheless, when MR images have diffuse edges or multiple similar objects (e.g. bladder close to prostate) within close proximity, such approaches have proven to be unsuccessful. In order to mitigate these problems, we proposed a new framework for bi-stage contours delineation algorithm based on directional active contours (DAC) incorporating prior knowledge of the prostate shape. We first explicitly addressed the prostate contour delineation problem based on fast globally DAC that incorporates both statistical and parametric shape prior model. In doing so, we were able to exploit the global aspects of contour delineation problem by incorporating a user feedback in contours delineation process where it is shown that only a small amount of user input can sometimes resolve ambiguous scenarios raised by DAC. In addition, once the prostate contours have been delineated, a cost functional is designed to incorporate both user feedback interaction and the parametric shape prior model. Using data from publicly available prostate MR datasets, which includes several challenging clinical datasets, we highlighted the effectiveness and the capability of the proposed algorithm. Besides, the algorithm has been compared with several state-of-the-art methods.

Quantifying and minimising systematic and random errors in X-ray micro-tomography based volume measurements

X-ray micro-tomography (XMT) is increasingly used for the quantitative analysis of the volumes of features within the 3D images. As with any measurement, there will be error and uncertainty associated with these measurements. In this paper a method for quantifying both the systematic and random components of this error in the measured volume is presented. The systematic error is the offset between the actual and measured volume which is consistent between different measurements and can therefore be eliminated by appropriate calibration. In XMT measurements this is often caused by an inappropriate threshold value. The random error is not associated with any systematic offset in the measured volume and could be caused, for instance, by variations in the location of the specific object relative to the voxel grid. It can be eliminated by repeated measurements. It was found that both the systematic and random components of the error are a strong function of the size of the object measured relative to the voxel size. The relative error in the volume was found to follow approximately a power law relationship with the volume of the object, but with an exponent that implied, unexpectedly, that the relative error was proportional to the radius of the object for small objects, though the exponent did imply that the relative error was approximately proportional to the surface area of the object for larger objects. In an example application involving the size of mineral grains in an ore sample, the uncertainty associated with the random error in the volume is larger than the object itself for objects smaller than about 8 voxels and is greater than 10% for any object smaller than about 260 voxels. A methodology is presented for reducing the random error by combining the results from either multiple scans of the same object or scans of multiple similar objects, with an uncertainty of less than 5% requiring 12 objects of 100 voxels or 600 objects of 4 voxels. As the systematic error in a measurement cannot be eliminated by combining the results from multiple measurements, this paper introduces a procedure for using volume standards to reduce the systematic error, especially for smaller objects where the relative error is larger.

Extraction of target specimens from bioholographic images using interactive graph cuts

It is necessary to extract target specimens from bioholographic images for high-level analysis such as object identification, recognition, and tracking with the advent of application of digital holographic microscopy to transparent or semi-transparent biological specimens. We present an interactive graph cuts approach to segment the needed target specimens in the reconstructed bioholographic images. This method combines both regional and boundary information and is robust to extract targets with weak boundaries. Moreover, this technique can achieve globally optimal results while minimizing an energy function. We provide a convenient user interface, which can easily differentiate the foreground/background for various types of holographic images, as well as a dynamically modified coefficient, which specifies the importance of the regional and boundary information. The extracted results from our scheme have been compared with those from an advanced level-set-based segmentation method using an unbiased comparison algorithm. Experimental results show that this interactive graph cut technique can not only extract different kinds of target specimens in bioholographic images, but also yield good results when there are multiple similar objects in the holographic image or when the object boundaries are very weak.

Multiple pedestrians tracking algorithm by incorporating histogram of oriented gradient detections

The authors propose an effective algorithm for multiple pedestrians tracking, which is constructed in the framework of particle filtering, and it is based on the combination of online boosting tracker and the histogram of oriented gradient (HOG) descriptor for human detection. The combination for the detector and tracker lies on following aspects. First, each detection result is associated to a tracker implemented by the online boosting, which gives the authors scheme robustness for multiple similar objects and then, the output of support vector machine classifier based on HOG is dynamically fused as a component in the observation metric in particle filtering, which makes the tracker more accurate in some difficult conditions. Finally, the states of some particles are replaced by the state given by the detector, so that the tracker can recover from failure quickly. Experiments show the effectiveness of their scheme.

Shape priors extraction and application for geodesic distance transforms in images and videos

We present a shape prior embedded geodesic distance transform for image and video segmentation. Whilst the existing segmentation algorithms have achieved impressive performance in many examples, they may fail in cases where the quality of the likelihood images is not satisfactory, or where multiple similar objects are in close proximity to one another. To deal with these problems, we embed shape prior knowledge in an image segmentation algorithm based on geodesic distance transform. Different from other segmentation methods, the proposed geodesic distance transform morphology operators consider three factors simultaneously: the geometric distance, weighted gradients, and the distance to the boundary of shape priors. As a result, it provides segmentation in line with the real shape of a particular kind of object. We also propose an effective shape prior extraction method that compute shape priors automatically. The proposed algorithm demonstrates positive results for many challenging images and video sequences in our experiments.

Object tracking based on particle filter with discriminative features

This paper presents a particle filter-based visual tracking method with online feature selection mechanism. In color-based particle filter algorithm the weights of particles do not always represent the importance correctly, this may cause that the object tracking based on particle filter converge to a local region of the object. In our proposed visual tracking method, the Bhattacharyya distance and the local discrimination between the object and background are used to define the weights of the particles, which can solve the existing local convergence problem. Experiments demonstrates that the proposed method can work well not only in single object tracking processes but also in multiple similar objects tracking processes.

The Effect of Proximity, Tall Man Lettering, and Time Pressure on Accurate Visual Perception of Drug Names

The aim of this study is to assess the effect of proximity and time pressure on accurate and effective visual search during medication selection from a computer screen. The presence of multiple similar objects in proximity to a target object increases the difficulty of a visual search. Visual similarity between drug names can also lead to selection error. The proximity of several similarly named drugs within a visual field could, therefore, adversely affect visual search. In Study 1, 60 nonpharmacy participants selected a target drug name from an array of mock drug packets shown on a computer screen, where one or four similarly named nontargets might be present. Of the participants, 30 completed the task with a time constraint, and the remainder did not. In Study 2, the same experiment was repeated with 28 pharmacy staff. In Study 1, the proximity of multiple similarly named nontargets within the specified visual field reduced selection accuracy and increased reaction times in the nonpharmacists. Time constraint also had an adverse effect. In Study 2, the pharmacy participants showed increased reaction times when multiple nontargets were present, but the time constraint had no effect. There was no effect of Tall Man lettering. The presence of multiple similarly named medications in close proximity to a target medication increases the difficulty of the visual search for the target. Tall Man lettering has no impact on this adverse effect. The widespread use of the alphabetical system in medication storage increases the risk of proximity-based errors in drug selection.

A TEAM OF HUMANOID GAME COMMENTATORS

We present a team of two humanoid robot commentators for AIBO robot soccer games. The two humanoids stand by the side lines of the playing field, autonomously observe the game, wirelessly listen to a "game controller" computer, recognize events, and select announcing actions that may require coordination with each other. Given the large degree of uncertainty and dynamics of the robot soccer games, we further introduce a "Puppet Master" control that allows humans to intervene, prompting the robots to commentate an event if previously undefined or undetected. The robots recognize events based on input from these three sources, namely own and shared vision, game controller, and occasional Puppet Master. We present the two-humanoid behavioral architecture and the vision-based event recognition, including a SIFT-based vision processing algorithm that allows for the detection of multiple similar objects, such as the identical shaped robot players. We introduce the commentating algorithm that probabilistically selects a commentating action from a set of weighted actions corresponding to a detected event. The probabilistic selection uses the game history and updates the action weights to effectively avoid repetition of comments to enable entertainment. Our work, corresponding to a fully implemented system, CMCast, with two QRIO robots, contributes a team of two humanoids fully executing a challenging observation, modeling, coordination, and reporting task.

Attentional resolution

Attention can enhance selectively the visual information processing of particular locations or objects. Recent studies have shown that this enhancement has limited spatial resolution, the smallest regions that can be isolated by attention are much coarser than the smallest details that can be resolved by vision. Multiple similar objects spaced more finely than the limit of attentional resolution cannot be individuated for further processing and can only be perceived as a grouped texture. As a result, at any given time, only part of the spatial and temporal information registered by the early sensory systems is available to conscious perception. It is likely that attentional resolution is limited at a stage beyond V1 and that it has a finer grain in the lower visual field than in the upper field. The spatial aperture of attention is elongated along the radial axis relative to fixation. The briefest temporal window of attention is also much broader than visual temporal resolution. Many perceptual phenomena related to rapid serial visual presentation may reflect the limited temporal resolution of attention.