Search For Objects Research Articles

Search for Visual Objects by Request in the Form of a Cluster Representation for the Structural Image Description

Search for Visual Objects by Request in the Form of a Cluster Representation for the Structural Image Description

Searching for compact objects in the single-lined spectroscopic binaries of the young Galactic cluster NGC 6231

Searching for compact objects in the single-lined spectroscopic binaries of the young Galactic cluster NGC 6231

Assessment of the deep oil and gas bearing potential onshore in the west of Azerbaijan (Tarsdallar area)


 
 
 The Tarsdeller oil and gas field, which has a brachyoanthicline structure, is the most promising area of the oil-gas region (OGR) in the Kura-Gabirri interfluve. Goal: Based on the reservoir characteristics of rock samples obtained from exploration wells drilled in the area adjacent to the field, petrophysical models have been compiled, reflecting the spatial variation in porosity and permeability of oil-containing reservoirs. Objects: Consideration of the petrophysical properties of core samples taken in the study area made it possible to confirm the oil and gas potential of deep-seated reservoirs. Methods: This analyzed the values and variation of petrophysical properties by depth, age and various physical factors. The observed wide range of changes in the reservoir properties of rocks in the study area is mainly due to tectonic changes, lithological heterogeneity of the sedimentary complex, differences in the depth of bedrocks, as well as the complexity of tectonic conditions. As a result, to predict the oil and gas content in deep reservoirs of the same structures, it is advisable to use the methods of exploration geophysics, as well as the results of changes inthe reservoir characteristics of rocks, determined by petrophysical data. At the same time, it is most expedient to focus on the study of Paleogene-Cretaceous tectonics for the search for oil and gas objects using prospecting and exploration (preferably aerial-photo-space) works at the field. Results: Models made in 3D format illustrate the predicted oil and gas potential and patterns of distribution of productive horizons involved in the geological structure of oil and gas deposits, along the depths and stratigraphic units. It is shown that although the permeability of the rocks in the area is low, the porosity values are favorable for the industrial accumulation of hydro-carbons. Fractured carbonate reservoirs of the Upper Cretaceous are the most promising among the Mesozoic sediments in the OGRbetween the Kura and Gabirri rivers. Deep-seated structural uplifts complicated by large-amplitude fractures are recommended as the priority targets for further geological exploration. However, the real potential of the Eocene deposits, which are widespread in the NGR between the Kura and Gabirri rivers, remains unexplored by deep drilling. This is due, on the one hand, to the incomplete opening of the Eocene section in the structures prepared for deep drilling, structural-tectonic and lithofacies features, as well as insufficient knowledge of drilling for oil and gas, and on the other hand, poor development of promising intervals of the section.
 
 

Semantic Regions of Interest Applied to Adaptive Image Compression, Color Enhancement, and Automated Pan-and-Scan

One of the most challenging problems in computer vision and image processing is the detection of the semantic regions of interest (SRoIs). In this article, we propose a method using OpenAI’s contrastive language-image pretraining (CLIP) model to find SRoIs by performing a semantic search for objects in the image that are detected by an object detection model called generic RoI extractor (GRoIE). Finding the SRoI can be used in different image-processing tasks such as image and video compression, enhancement, and reformatting. By knowing the SRoI within images, we can improve the visual quality of images by compressing the more important parts with higher quality and the less important parts, such as the background, with a lower quality. This operation can be achieved without changing the overall compression ratio and the peak signal-to-noise ratio (PSNR) quality metric. Finding the SRoI can make the processes of image enhancement and color correction more accurate by focusing only on the important parts. Moreover, in the image reformatting process, the important parts of the image may be lost. But by using the SRoI, we can reformat the image in a better way by keeping the regions within a frame that draw the greatest audience attention.

Efficient Person Search: An Anchor-Free Approach

Person search aims to simultaneously localize and identify a query person from uncropped images. To achieve this goal, state-of-the-art models typically add a re-id branch upon two-stage detectors like Faster R-CNN. Owing to the ROI-Align operation, this pipeline yields promising accuracy as re-id features are explicitly aligned with the corresponding object regions, but in the meantime, it introduces high computational overhead due to dense object anchors. In this work, we present an anchor-free approach to efficiently tackling this challenging task, by introducing the following dedicated designs. First, we select an anchor-free detector (i.e., FCOS) as the prototype of our framework. Due to the lack of dense object anchors, it exhibits significantly higher efficiency compared with existing person search models. Second, when directly accommodating this anchor-free detector for person search, there exist several misalignment issues in different levels (i.e., scale, region, and task). To address these issues, we propose an aligned feature aggregation module to generate more discriminative and robust feature embeddings. Accordingly, we name our framework as Feature-Aligned Person Search Network (AlignPS). Third, by investigating the advantages of both anchor-based and anchor-free models, we further augment AlignPS with an ROI-Align head, which significantly improves the robustness of re-id features while still keeping our model highly efficient. Our framework not only achieves state-of-the-art or competitive performance on two challenging person search benchmarks, but can be also extended to other challenging searching tasks such as animal and object search. All the source codes, data, and trained models are available at: https://github.com/daodaofr/alignps .

The influence of the stimulus category on the characteristics of eye movements in visual search

Introduction. The speed and efficiency of visual search are important characteristics of the activities of a significant range of specialists working with visual information. Specialists have carefully studied the features of visual search from the point of view of cognitive psychology and in connection with the modeling of oculomotor activity. It has been shown that its effectiveness and the parameters of search eye movements are influenced by both stimulation factors and some cognitive factors (features of the subject's preset, the influence of previous search stages, etc.). We assume that the search features are also related to how well a person represents the desired stimulus: whether it belongs to a familiar semantic category, how accurate and unambiguous its visual representation is.
 The study aims to explore the relationship between the parameters of eye movements in visual search and the features of the mental representation of the target object.
 Materials and methods. We have created sets of stimuli belonging to different semantic categories, characterized by different accuracy of mental representations, but having similar sensory qualities. Each object (both the target stimulus and the distractor) in the search field was a black circle on a white background with four black segments inside. The segments formed one of three configurations: a) a face; b) a Roman numeral; c) a random configuration. The study participants were asked to find in a matrix of 84 objects a specific target stimulus belonging to one of the three specified categories. The scientists recorded the participant's eye movements during the search.
 Results. The results of the study generally confirmed the hypothesis about the relationship of search characteristics with the category of the target object. The smallest number of errors, the minimum duration of fixation and the maximum amplitude of saccades characterized the search for Roman numerals — objects with a well-formed and well-defined visual representation. The most difficult, as expected, was the search for objects representing random configurations of segments.
 Limitations. Since the study was largely exploratory in nature, the results obtained require clarification on a wider sample. It is possible that the characteristics of the search were influenced by such a factor not controlled in the study as the configuration of the lines, namely, the presence or absence of their intersections, which should be taken into account in further work.
 Conclusions. The better the visual representation of the target object is formed, the more successful its search is. In this case, eye movements are characterized by a shorter duration of fixation and a higher frequency of saccades of greater amplitude.
 Ethics. The study was conducted in compliance with the ethical principles set out in the Helsinki Declaration of the World Medical Association, all participants gave voluntary informed consent.

Hydroacoustic Bottom Soil Profiling Device with Synthetic Aperture

Introduction. Acoustic profiling is a conventional method for studying the geological structure of the seabed. To this end, a low-frequency acoustic profiler with an operating frequency range of 1…14 kHz can be used. However, under lower operating frequencies, the difficulty of achieving the required resolution arises. The problem of improving the angular resolution of a hydroacoustic device, particularly in the direction of the carrier movement, remains to be a priority task in the search and detection of objects on the seabed, as well as the study of the bottom soil structure. Angular resolution can be improved through several approaches, including an algorithm for synthesizing the antenna aperture based on the law of the phase change of the reflected signal. This approach is relevant in the design of highresolution hydroacoustic tools.Aim. To demonstrate the possibility of constructing a bottom soil profiling device, as well as the possibility of increasing its angular resolution based on an algorithm for synthesizing the antenna aperture of an acoustic profiler.Materials and methods. The study employed the groundwork data obtained for side-scan sonars with a synthetic aperture in terms of constructing an antenna device and the methods of radio-wave excitation developed for radar systems.Results. The possibility of synthesizing the antenna aperture for an acoustic profiler of the bottom soil was studied. An algorithm for synthesizing the antenna aperture was investigated along with phase distortions of the trajectory signal and their influence on the sonar image. The fundamental principles of processing the trajectory signal were considered.Conclusion. An approach to increasing the resolution of an acoustic profiler when performing search and survey tasks is proposed. The proposed design of the acoustic part of a profiling device can be used in the development of search and survey hydroacoustic devices with a high angular resolution.

Bibliometric Analysis of Determining Factors in Entrepreneurial Intention

In entrepreneurship, its various theories and elements gain importance in the academic community for the orientation in the professional and entrepreneurial training processes that are being carried out by higher education institutions HEIs. Therefore, this paper provides a systematic review of the published literature based on the determinants of entrepreneurial intention. To that end, an analysis was conducted using bibliometric and network indicators within the Scopus data source, which was restricted to objective search criteria. The main findings show that, since 1943, certain success factors such as: attitude, learning, behavior, economic growth, education, value creation, incubation resources, personality traits, and entrepreneurial intentions have been linked.
 
 Received: 23 November 2022 / Accepted: 22 January 2023 / Published: 5 March 2023

Labeling Out-of-View Objects in Immersive Analytics to Support Situated Visual Searching.

Augmented Reality (AR) embeds digital information into objects of the physical world. Data can be shown in-situ, thereby enabling real-time visual comparisons and object search in real-life user tasks, such as comparing products and looking up scores in a sports game. While there have been studies on designing AR interfaces for situated information retrieval, there has only been limited research on AR object labeling for visual search tasks in the spatial environment. In this article, we identify and categorize different design aspects in AR label design and report on a formal user study on labels for out-of-view objects to support visual search tasks in AR. We design three visualization techniques for out-of-view object labeling in AR, which respectively encode the relative physical position (height-encoded), the rotational direction (angle-encoded), and the label values (value-encoded) of the objects. We further implement two traditional in-view object labeling techniques, where labels are placed either next to the respective objects (situated) or at the edge of the AR FoV (boundary). We evaluate these five different label conditions in three visual search tasks for static objects. Our study shows that out-of-view object labels are beneficial when searching for objects outside the FoV, spatial orientation, and when comparing multiple spatially sparse objects. Angle-encoded labels with directional cues of the surrounding objects have the overall best performance with the highest user satisfaction. We discuss the implications of our findings for future immersive AR interface design.

Increase of the Automotive Power Transistor Modules Manufacture Reliability Using Ai Detecting System for Soldering Splashes

This study discusses the automated visual inspection of electronic boards used in the mass production of power electronics equipment. Soldering splashes, produced during the necessary manufacturing stages, can affect the hybrid power semiconductor modules' quality, specifications and lifespan. Splashes from soldering may appear in some electronic boards areas when they may cause decreasing of quality of these boards and need to be removed. Image analysis algorithms are used to search for such areas. The automated inspection is based on neural network YOLO. The images of electronic boards, acquired by authors in SEMIKRON Slovakia company, were used as training, testing and validation dataset for the YOLO network. The custom recording system was designed to acquire those images. The implementation of this method may increase the object search success and then thereby increase the overall quality of module production.

Effect of display of YOLO’s object recognition results to HMD for an operator controlling a mobile robot

An operator feels a burden when he/she controls a rescue robot remotely because he/she has to keep watching camera images to find the target object. We think that this burden can be reduced by the combination of Head Mounted Display (HMD) and object recognition by deep learning. In the first half part of this study, we examine the effect that how presentation method by You Only Look Once (YOLO), a deep learning algorithm, and its recognition results to an operator wearing HMD. In the experiment, three methods of presentation were set: no display of object recognition, display only one object recognition result, and display 80 kinds of object recognition results. Under each presentation method, we measured the time it took for the operator to operate the robot and complete the given task. Additionally, we ask a questionnaire for each experiment. The results of the questionnaire showed that the method to present only one object recognition result was useful. In the second half part of this study, we develop a system to present 3D images with YOLO added, to further ease the burden of object search. Furthermore, we numerically prove that this system represents depth. In the experiment, two methods of displaying were set up: 2D images with Bounding Box (BB) by YOLO and 3D images with BB by YOLO. For each method of presentation, the operator operated the robot and recorded the number of objects found within a time limit. Additionally, we asked a questionnaire at the end of the search in each condition and at the end of all the experiments. The results of the questionnaire suggested points that need to be improved. Furthermore, we consider the flicker of the image found in the experiment.

Semantic Grounding for Long-Term Autonomy of Mobile Robots Toward Dynamic Object Search in Home Environments

In this article, we presents an effective semantic grounding scheme to long-term mobile robots for dynamic object search in open/dynamic home environments. The challenge of this strategy lies in dealing with situations where the robot suffers the home environment with an unknown layout of the object and the target object is moved without the robot’s knowledge. Therefore, a general probabilistic model and a semantic model are established by investigating typical spatial location relations between object and room type, in order to guide the robot to prioritize the search effort for spaces that are most promising to find the target object. Furthermore, a semantic grounding solution is proposed to address the mutual independence between objects, endowing the robot with inference capability for dynamic object search. Also, our proposal can be shared with other robots that have not explored the environment in order to perform object search tasks. The presented scheme is fully evaluated by extensive experiments in the real-world environment. The results demonstrate the validity of our approach by comparing it with the other three schemes in terms of search time and trajectory length, and show that the proposed method allows the mobile robot to efficiently and robustly find the dynamic object while achieving the human-like performance.

A reduction procedure and pipeline for the detection of trans-Neptunian objects using occultations

Abstract Kuiper belt objects smaller than a few kilometres are difficult to observe directly. They can be detected when they randomly occult a background star. Close to the ecliptic plane, each star is occulted once every tens of thousands of hours, and occultations typically last for less than a second. We present an algorithm, and companion pipeline, for detection of diffractive occultation events. Our approach includes: cleaning the data; an efficient and optimal matched filtering of the light curves with a template bank of diffractive occultations; treating the red-noise in the light curves; injection of simulated events for efficiency estimation; and applying data quality cuts. We discuss human vetting of the candidate events in a blinded way to reduce bias caused by the human-in-the-loop. We present Markov Chain Monte Carlo tools to estimate the parameters of candidate occultations, and test them on simulated events. This pipeline is used by the W-FAST. The methods discussed here can be applied to searches for other Trans-Neptunian objects, albeit with larger radii that correspond to a larger diffraction length scale.

Visual servoing of quadrotor UAVs for slant targets with autonomous object search

In this paper, an enhanced visual servoing method is designed for a quadrotor unmanned aerial vehicle (UAV) based on virtual plane image moments, under underactuation and tight coupling constraints of UAV kinematics. Moreover, in order to make the UAV search visual targets autonomously in target vicinity during flight, a flexible flight system is developed with stages of take-off, target searching, and image-based visual servoing (IBVS). With dual-camera sensor configuration, the UAV system searches targets from given directions while making localization. A virtual image plane is constructed and image moments are adopted to decouple UAV lateral movement. For a non-horizonal target, homography is utilized to construct the target plane and transform it into a horizonal plane. Backstepping techniques are used to derive the nonlinear controller to realize the IBVS strategy. Stability analysis proves global asymptotic performance of the closed-loop system. Experimental verification shows feasibility of the overall flight system and effectiveness of the visual servoing controller.

Semantic field SEEK: Shughni data in the light of lexical typology

The paper deals with verbs and verbal constructions with the meaning ‘to seek, to search, to look for’ (the semantic field SEEK) in the Shughni language against a typological background. On the basis of dictionary data and the results of elicitation sessions with native speakers, the peculiarities of the lexicalization of the SEEK field in modern Shughni are revealed, and assumptions are made regarding the dynamics of the development of this system from the middle of the twentieth century to the present day. We show that, in this semantic field, there is a clear lexical dominant — the verb x̌ikīdow — with a very wide scope of use, including search for objects of different types (specific and non-specific) by different subjects (humans or animals) and in different ways (visually, tactually, with the help of speech activity, etc.). In some semantic domains within the field of seeking, it competes with other verbs and constructions: in the domain of searching a space, the construction tīr či bīr čīdow ‘to turn upside down’ is used, as well as the verbs čīx̌tow ‘to look’ and qilāptow ‘to rummage’, denoting visual and tactual search, respectively; the construction siroq čīdow (lit. ‘search to do’) can describe a search carried out by asking others around. Several verbs denoting, according to the dictionary data, a search in the dark (warθāptow, zi(r)ɣāptow, and warwarθtow), went out of use, two more — tilāptow ‘to ask’ and x̌iqāptow ‘to chat, invent’ — lost the meaning of seeking. The description of the system of the Shughni verbs of seeking is supplemented by an analysis of their polysemy: cases of using these verbs outside the SEEK field are highlighted and the nature of such colexifications is discussed.

Integrating Semantic Awareness and Probabilistic Priors for Object Search in Indoor Environments

Integrating Semantic Awareness and Probabilistic Priors for Object Search in Indoor Environments

ANALYSIS OF METHODS AND DEVELOPMENT OF THE CONCEPT OF GUARANTEED DETECTION AND RECOGNITION OF EXPLOSIVE OBJECTS

The subject of the article are the methods of detection and recognition of explosive objects. The aim of the work is to develop the main provisions of the concept of guaranteed detection and recognition of explosive objects. The following tasks were solved in the article: an analysis of existing approaches to the use of traditional single and combined, as well as non-traditional (biological) methods of detecting explosive objects, development of a classification table of methods of detection of explosive objects according to physical principles, analysis of advantages and disadvantages of the considered methods of detection of explosive objects, development of comparative table methods of detecting explosive objects, formulation of the introductory provisions of the concept of guaranteed detection and recognition of explosive objects. The following methods are used – methods of comparison, methods of abstraction, methods of analysis and synthesis, methods of scientific induction. The following results were obtained – an analysis of the features of the existing traditional and non-traditional (biological) methods of detecting explosive objects was carried out. A classification of methods for detecting explosive objects is proposed, taking into account the parameters that affect the probability of detection and productivity. The results of a comparative analysis of explosive object detection methods are presented in tabular form according to the following indicators: type of interaction with explosive objects, platform type, potential productivity, information technology support, quality parameters, and economic indicators. The main provisions of guaranteed detection and recognition of explosive objects are formulated. Conclusions: the low productivity of the existing methods does not allow for quick and effective clearing of the territory contaminated by explosive objects, which leads to a large number of injuries and deaths of people due to the detonation of explosive objects. The use of individual detection methods alone cannot significantly increase the probability of detecting explosive objects. To increase the productivity and safety of the search and disposal of explosive objects, it is advisable to use unmanned intelligent platforms to deliver information and measurement tools.

Do chimpanzees see a face on Mars? A search for face pareidolia in chimpanzees.

We sometimes perceive meaningful patterns or images in random arrangements of colors and shapes. This phenomenon is called pareidolia and has recently been studied intensively, especially face pareidolia. In contrast, there are few comparative-cognitive studies on face pareidolia with nonhuman primates. This study explored behavioral evidence for face pareidolia in chimpanzees using visual search and matching tasks. Faces are processed in a configural manner, and their perception and recognition are hampered by inversion and misalignment of top and bottom parts. We investigated whether the same effect occurs in a visual search for face-like objects. The results showed an effect of misalignment. On the other hand, consistent results were not obtained with the photographs of fruits. When only the top or bottom half of the face-like object was presented, chimpanzees showed better performance for the top-half condition, suggesting the importance of the eye area in face pareidolia. In the positive-control experiments, chimpanzees received the same experiment using human faces and human participants with face-like objects and fruits. As a result, chimpanzees showed an inefficient search for inverted and misaligned faces and humans for manipulated face-like objects. Finally, to examine the role of face awareness, we tested matching a human face to a face-like object in chimpanzees but obtained no substantial evidence that they saw the face-like object as a "face." Based on these results, we discussed the extents and limits of face pareidolia in chimpanzees.

Geological model and prospects of oil and gas bearing deposits of the silurian complex of the Pechora sea shelf

The task of this work is the desire of the authors to draw attention to the problem of ac-celerating the pace of searches for promising Arctic objects on the shelf of the Pechora Sea and additional exploration of already discovered deposits in the Silurian oil and gas complex on the territory of the Varandei-Adzva structural zone (VASZ). The main result of the previously per-formed offshore exploration work is the identification of a new oil-bearing area in the eastern part of the Pechoromorsky shelf. Large oil fields have been discovered here: Prirazlomnoye, Varandey-sea, Medynskoye-sea, Dolginskoye. The proven reserves and resources of hydrocarbons (HC) allow us to assume that in the near future a new oil-producing area will be formed on the shelf of the Pechora Sea. In total, the initial raw oil resources in the Pechora Sea are more than 3500 mln tons. Keywords: Pechora sea; silurian oil and gas complex; silurian; oil.

Analysis of modern methods of search and classification of explosive objects

The article is devoted to the analysis of existing methods of searching for explosive objects on the surface of the earth and under it, and to the development of new effective approaches to solving the problem. We focus on developing solutions based on AI technologies and methods that use publicly available hardware, structural methods, and machine learning methods The problems and their solutions mentioned in the article are quite specific and, despite the relevance of the topic of searching for explosive objects, poorly developed. The main reason for this situation is either the lack of information in the public domain, when developments are carried out by military departments or private companies, or the relatively low development of countries that suffer from the problem of demining territories where military operations have been or are being conducted. From 2014 to 2022, on the territory of Ukraine, the area affected by explosive objects was approximately equal to the area of Croatia, which took 20 years to clear the territories after the war in the Balkans (1991–1995). After 2022, the territory affected by explosive objects increased several times. The intensity of shelling can currently be compared to the First and Second World Wars, so it is safe to say that the problem of finding explosive objects has reached a higher level. Therefore, considering the volume of data and the scale of the affected territories, we decided to study the main directions and modern methods of searching and classifying explosive objects in order to use them to create a system or a framework for solving the given task. The results of this article are planned to be used to create a single algorithmic environment for solving the problems of finding explosive objects, which, if necessary, will be able to process data from different sources of information, with different degrees of detail and depth.