Automatic Capture Research Articles

Software Complex for Documenting Graphical Applications Based on Topological Decomposition

Contemporary software products and services necessitate extensive accompanying information, including user manuals and documentation. The task becomes more complex in the case of graphical applications, as it requires capturing screenshots and highlighting the described elements. Attaining high-quality visual information using conventional means demands considerable effort and time, since, unlike text, raster images do not provide an interface for retrieving metadata about individual elements, necessitating programmatic extraction of such information. This study presents a software complex designed to automate and expedite this process. The primary focus is on a topo­logical decomposition method that segments images based on topological features such as pixel brightness and proximity, enabling the segmentation of screenshots into control elements. This approach demonstrates high accuracy (96-99 %) in identifying graphical components across most tested images. The software product encompasses functions for the automatic capture of application windows, an algorithm for identifying control elements, and the capability to export the entire image or a specified region. Additionally, an algorithm for generating output templates for image insertion has been implemented. The decomposition method has exhibited effectiveness across the majority of test data. In instances with shadow effects, noise levels were recorded at a maximum of 34 %; however, in most cases, this figure did not exceed 3 %. Further research may contribute to enhancing these results. The findings of this study could prove valuable for the continued exploration of graphical user interface decomposition and the assessment of the method's effectiveness on diverse images. The developed software complex significantly simplifies and accelerates the documentation process, thereby opening new avenues in the automation of software development.

REMOTE INTELLIGENT ROBOT MANIPULATOR CONTROL SUBSYSTEM

The issue of expanding the functionality of an intelligent software and hardware subsystem for capturing an object by a robot manipulator was considered. It is shown that the task of expanding the functionality of the subsystem for free-form object capture by a robot manipulator with technical vision based on the operations of building a depth map from the obtained stereo image is insufficiently developed and can be solved by integrating a remote robot cntrol subsystem connected to a local network or the Internet into the robotic complex. The subsystem was developed using a Raspberry Pi microcomputer, a Stereo Pi board, a board for connecting eight Adafruit PCA9865 servos, Raspberry PI cameras, and a camera module. The structure and electrical diagram of the subsystem, a description of the software, and the main code snippet of the web application program are presented. Based on the analysis of the research results, it is established that the proposed remote control subsystem provides an extension of the functionality of the intelligent software and hardware subsystem for capturing an object by a robot manipulator, since in addition to the functions of automatic object capture, due to remote adjustment of the accuracy and manufacturability of object capture, it provides the functions of high-precision positioning of objects relative to each other and especially the effort on the robot manipulator to perform precise operations. The reliability of the object capture is also increased due to the fact that the subsystem allows for the additional involvement of a team of competent users. The proposed system increases the degree of automation, provides the possibility of moderation, collective administration, and user feedback during robot operation.

Automatic Detection and Tracking of Objects of Interest in Video Data with Global Motion

Introduction. At present, automatic capture and tracking of moving objects in video data obtained by a video camera mounted on a mobile carrier represents a relevant research task. Its successful solution is challenged by such factors, as a non-uniform background, object overlapping between one another and the background, significant and rapid changes in the size of the object of interest, abrupt changes in the movement trajectory of the mobile carrier.Aim. To develop an automatic method for detecting moving objects followed by their tracking in video data obtained under difficult observation conditions. An additional requirement imposed on the tracking stage consists in the restriction of computing resources.Materials and methods. The method is based on a convolutional neural network with a YOLO architecture. Due to the restriction of computing resources, object tracking is implemented without neural network solutions. In order to ensure stable tracking, several detectors are used simultaneously with the subsequent analysis of the data obtained. The tracking stage involves a detector based on histograms of oriented gradients (HOG), supplemented by a detector based on correlation filtering and motion trajectory prediction based on the Kalman filter.Results. At the automatic detection stage, the TPR, averaged over all video files participating in the experiments, was equal to 0.81, with the FPR corresponding to 0.10. At the tracking stage, the failure rate (tracking failures) was 6·10 –5.Conclusion. The proposed method can be successfully used to detect and track objects at a distance of 1500 m with an object projection size on the frame of 5 × 5 pixels under the conditions of global motion, a non-uniform background, and significant changes in the properties of the object of interest.

Autonomous positioning, capturing, and grasping mechanism for robot end effectors based on the attraction domain relationship

Purpose This paper aims to solve the problem of uncertain position and attitude between unstructured terrain robot and grasped target and insufficient control accuracy in extreme environment, a grasping mechanism based on attraction domain relationship is proposed, which can realize autonomous positioning, capturing and grasping of robot under low control accuracy. Design/methodology/approach The grasping mechanism was designed, taking inspiration from fishing behavior this mechanism introduces attraction domains and flexible-elastic structures through the active and passive ends to achieve automatic positioning and capture. After the capture is completed, the grasping mechanism connects the active end and the passive end, simultaneously relying on the gravity of the target object to achieve locking and release between the robot and the target object. This paper adopts theoretical, simulation and experimental verification methods to conduct theoretical and simulation analysis on the autonomous positioning and grasping process of the mechanism, and produces grasping experimental prototypes with different positions and postures. Findings The experiment shows that the gripping mechanism designed in this paper can achieve automatic positioning capture and gripping of large deviation situations under low control accuracy, with a displacement deviation of up to 10 mm (about 1/6 diameter of the end of the mechanism) and an angle deviation of up to 3°. The scientific research task in the extremely high altitude environment has finally been successfully accomplished. Originality/value Inspired by fishing behavior, this paper proposes a positioning, capturing and grasping mechanism. The attraction area built with permanent magnets, coupled with the flexible connection, enables precise capture under low control, while the grasping mechanism can also rely on gravity to self-lock and release.

Emojis and affective priming in visual word recognition

ABSTRACT Emojis are frequently used in digital communication to supplement the lack of non-verbal cues, but their integration during reading has not been thoroughly examined. This study explores the interplay between language and emotion by testing the influence of emotional valence and face-status of emojis on visual word recognition. Two online experiments were conducted with 92 native English-speaking university students, examining priming effects between congruent (e.g. delicious) and incongruent (e.g. hate) prime-target pairs, varying the face-status of the emoji prime (face vs. non-face) and the valence (positive vs. negative) of the word target. Irrespective of valence, face emojis demonstrated a processing advantage over non-face emojis, implying automatic attention capture. Additionally, the results revealed an interaction between prime-target congruency and valence, with a facilitatory effect for positive, but not negative, items, suggesting a valence-specific mechanism of affective priming in the lexical decision task. The research suggests that the rapid integration of emoji content occurs during the early stages of visual word recognition, with heightened attentional sensitivity to both face-like and positive stimuli when reading digital communications.

6869 Endocrinologist Perceptions of the Value of Digital Health Tools to Assist in the Management of Insulin Dosing

Abstract Disclosure: M.B. Davidson: Employee; Self; Eli Lilly & Company. B. Liao: Employee; Self; Eli Lilly & Company. C. Heckman: Employee; Self; Eli Lilly & Company. Background and aims: Insulin management is complex for both people with diabetes (PwD) and the healthcare providers (HCP) who treat them. Digital tools, such as connected insulin pen (CIP) platforms may reduce burden and improve suboptimal insulin dosing, especially through automatic capture of insulin dose events and integration with glucose values. However, the acceptance and utilization of CIP platforms in clinical practice remains lower than expected, given this clinical care gap. This study attempted to ascertain beliefs of HCP leading to this finding. Methods: For this cross-sectional study, 190 members of The Endocrine Society who self-identified as “clinical endocrinologists who treat diabetes” completed a 10-question survey on their perception of digital health tools and their value in clinical practice. Results: In their patients using multiple daily injections (MDI), defined as three or more injections of insulin per day, documentation of insulin dosing data was identified as “important” or “very important’ to 81% of respondents. However, only 48% of respondents are “confident” or “very confident” in the reliability of manually captured insulin and glucose data. While 92% of respondents indicated that a digital platform that displays insulin and glucose data would be “beneficial” or “very beneficial” for making dosing recommendations, 60% have found that a CIP platforms had been “beneficial” or “very beneficial” in helping PwD improving their glycemic control. Conclusion: A group of almost 200 clinician members of the Endocrine Society, who actively treat PwD, indicated that in their PwD using MDI regimens, access to combined insulin and glucose data is important, and digital capture of this data would be beneficial. However, this data is often unavailable or unreliable. The survey revealed a gap between their belief that availability of glucose data and insulin dosing data can help them make dosing recommendations and their belief that these will translate into impacting glycemic control of patients. Further research would be helpful to determine how diabetes technology may better support these HCP and the PwD they treat in practice. Presentation: 6/1/2024

ATD-300 an AI empowered miniature acoustic device for automatic gunshot attacks detection and image capture

Built on over 30 years of acoustic source recognition expertise in transient impulsives signature, recently highly empowered by neural networks, and wave propagation model driven algorithms, ACOEM has developed the ATD-300, an acoustic sensor for automatic gunshot detection, identification and localization. What sets this sensor apart from the competition is its reliability in terms of detection rate over false alarm and robust high resolution shooter pointing capability. Therefore it can be used directly by police departments without third-party human validation. Resulting from the design effort to distribute intelligence among a collaborative sensor network, processing and artificial intelligence algorithms are fully embedded locally, enabling an alert to be raised in less than three seconds and a camera to be pointed at the shooter with only one sensor. In addition, as the system is already installed in several American cities, we will show how a device-to-server distributed artificial intelligence approach continuously improves the performance of the solution, fed by real field data.

Exploring the spatial interference effects elicited by social and non-social targets: A conditional accuracy function approach.

Recent studies employing the spatial interference paradigm reveal qualitative differences in congruency effects between gaze and arrow targets. Typically, arrows produce a standard congruency effect (SCE), with faster responses when target direction aligns with its location. Conversely, gaze targets often lead to a reversed congruency effect (RCE), where responses are slower in similar conditions. We explored this dissociation using the Conditional Accuracy Function (CAF) to assess accuracy across reaction time bins. Using a hierarchical linear mixed modelling approach to compare cropped eyes, and full faces as social stimuli, and arrows as non-social stimuli, we synthesized findings from 11 studies, which led to three distinct models. The results showed that with non-social targets, incongruent trials exhibited lower accuracy rates in the first bin than in subsequent bins, while congruent trials maintained stable accuracy throughout the distribution. Conversely, social targets revealed a dissociation within the fastest responses; alongside a general reduction in accuracy for both congruency conditions, congruent trials resulted in even lower accuracy rates than incongruent ones. These results suggest with gaze targets that additional information, perhaps social, in addition to the automatic capture by the irrelevant target location, is being processed during the earlier stages of processing.

Novel integrated system for power, hydrogen, and ammonia production using direct oxy-combustion sCO2 power cycle with automatic CO2 capture, water electrolyzer, and Haber-Bosch process

Renewable energy-based multigeneration systems for power, hydrogen, and ammonia production suffer from intermittency issues and high production costs. On the other hand, conventional fossil-fuel-based systems yield high levels of CO2 emissions. Thus, to eliminate the intermittency issue with low production costs and near zero CO2 emissions, a novel integrated, efficient, compact, and interactive system for power, hydrogen (H2), and ammonia (NH3) production is developed and investigated in this study. A direct oxy-combustion supercritical carbon dioxide (DOC-sCO2) power plant is integrated with a water electrolyzer (WE) and Haber-Bosch process (HBP) system for the cogeneration of power, H2, and NH3, respectively. The power consumption of the air separation unit (ASU) of the DOC-sCO2 cycle is significantly reduced by the utilization of the oxygen produced from the WE system. Moreover, the cost of ammonia production is minimized by the internal utilization of the nitrogen gas produced by the ASU. Furthermore, the feedwater of the WE system is supplied internally from the DOC-sCO2 power block. In addition, the CO2 emissions of the system are captured to realize near-zero emissions to the ambiance. Based on these features, thorough energy, exergy, economic, and environmental (4E) analyses were performed to evaluate the performance of the proposed system. The obtained results showed an impressive levelized cost of electricity of 3.72¢/kWh, levelized cost of hydrogen of 1.162$/kgH2, and ammonia of 0.197$/kgNH3 at an overall energy and exergy efficiencies of 44.09 %, and 63.6 %, respectively. The system has no emissions of nitrogen oxides (NOx) and reduces the CO2 emissions of the H2 production by 49.6 % and of the NH3 production by 48.8 % compared to the conventional gasification, SMR, and HBP systems, respectively. This research work also provides insights into wet versus dry cooling, part load operation, and the advantages over renewable-based systems in terms of stability, reliability, and economic feasibility.

Prevalence of Leber hereditary optic neuropathy in the Community of Madrid (Spain), estimation with a capture-recapture method

BackgroundLeber hereditary optic neuropathy (LHON) typically presents in young adults as bilateral painless subacute visual loss. Prevalence data are scarce. The aim of this study was to examine the validity of different ascertainment sources used in population-based rare diseases registries to detect cases, and to explore the impact of a capture-recapture method in the estimation of the prevalence of LHON in the Autonomous Community of Madrid (ACM) in 2022.MethodsDescriptive cross-sectional population-based study. Potential LHON cases were detected by automatic capture from the healthcare information sources usually explored for the Regional Registry for Rare Diseases (SIERMA). Ophthalmologists provided data from their clinical registry. Positive predictive values (PPV) and sensitivity with 95% confidence intervals (CI) were estimated. Global and by sex prevalences were calculated with confimed cases and with those estimated by the capture-recapture method.ResultsA total of 102 potential LHON cases were captured from healthcare information sources, 25 of them (24.5%) finally were confirmed after revision, with an overall PPV of 24.5% (95%CI 17.2–33.7). By source, the electronic clinical records of primary care had the highest PPV (51.2, 95%CI 36.7–65.4). The ophthalmologists clinical registry provided 22 cases, 12 of them not detected in the automatic capture sources. The clinical registry reached a sensitivity of 59.5% (95%CI 43.5–73.6) and the combination of automatic capture sources reached a 67.6% (95%CI: 51.5–80.4). The total confirmed cases were 37, with a mean age of 48.9 years, and a men: women ratio of 2.4:1. Genetic information was recovered in 27 cases, with the m.3460 mutation being the most frequent (12 cases). The global prevalence was 0.55 cases/100,000 inhabitants (95%CI 0.40–0.75), and with the capture-recapture method reached 0.79 cases/100,000 (95%CI 0.60–1.03), a 43.6% higher, 1.15 cases/100,000 (95%CI 0.83–1.58) in men and 0.43 cases/100,000 (95%CI 0.26–0.70) in women.ConclusionsThe prevalence of LHON estimated in the ACM was lower than in other European countries. Population-based registries of rare diseases require the incorporation of confirmed cases provided by clinicians to asure the best completeness of data. The use of more specific coding for rare diseases in healthcare information systems would facilitate the detection of cases. Further epidemiologic studies are needed to assess potential factors that may influence the penetrance of LHON.

The distinct effects of fearful and disgusting scenes on self-relevant face recognition

Self-face recognition denotes the process by which a person can recognize their own face by distinguishing it from another’s face. Although many research studies have explored the inhibition effect of negative information on self-relevant face processing, few researchers have examined whether negative scenes influence self-relevant face processing. Fearful and disgusting scenes are typical negative scenes, but little research to data has examined their discriminative effects on self-relevant face recognition. To investigate these issues, the current study explored the effect of negative scenes on self-relevant face recognition. In Study 1, 44 participants (20 men, 24 women) were asked to judge the orientation of a target face (self-face or friend-face) pictured in a negative or neutral scene, whereas 40 participants (19 men, 21 women) were asked to complete the same task in a fearful, disgusting, or neutral scene in Study 2. The results showed that negative scenes inhibited the speed of recognizing self-faces. Furthermore, the above effect of negative scenes on self-relevant face recognition occurred with fearful rather than disgusting scenes. Our findings suggest the distinct effects of fearful scenes and disgusting scenes on self-relevant face processing, which may be associated with the automatic attentional capture to negative scenes (especially fearful scenes) and the tendency to escape self-awareness.

Effect of Cognitive Processing Style on Attentional Blink during Analogue Trauma

Background: This study attempted to examine the impact of each cognitive processing style (bottom-up, top-down) on attention patterns following a traumatic experience by measuring attentional blink (AB).Methods: Participants were 37 university students with no direct experience of traffic accidents. They were randomly assigned to either a single-task (bottom-up processing) group or a dual-task (top-down processing) group, who performed the Rapid Serial Visual Presentation (RSVP) after watching an analogue trauma video. The correct response rate to target stimuli was then compared between groups based on stimulus type and stimulus presentation interval.Results: In the dual-task group, no significant differences were found. However, in the single-task group, a trend towards a decrease in the correct response rate to the target stimulus was observed 200 ms after the presentation of the traumatic stimulus (i.e., AB).Conclusions: This tendency can be explained by the phenomenon of automatic attentional capture by traumatic stimuli, suggesting a relationship between bottom-up processing and attentional bias.

The defect feature extraction of ultrasonic phased array detection based on adaptive region growth.

An ultrasonic phased array defect extraction method based on adaptive region growth is proposed, aiming at problems such as difficulty in defect identification and extraction caused by noise interference and complex structure of the detected object during ultrasonic phased array detection. First, bilateral filtering and grayscale processing techniques are employed for the purpose of noise reduction and initial data processing. Following this, the maximum sound pressure within the designated focusing region serves as the seed point. An adaptive region iteration method is subsequently employed to execute automatic threshold capture and region growth. In addition, mathematical morphology is applied to extract the processed defect features. In the final stage, two sets of B-scan images depicting hole defects of varying sizes are utilized for experimental validation of the proposed algorithm's effectiveness and applicability. The defect features extracted through this algorithm are then compared and analyzed alongside the histogram threshold method, Otsu method, K-means clustering algorithm, and a modified iterative method. The results reveal that the margin of error between the measured results and the actual defect sizes is less than 13%, representing a significant enhancement in the precision of defect feature extraction. Consequently, this method establishes a dependable foundation of data for subsequent tasks, such as defect localization and quantitative and qualitative analysis.

Linear Potentiometer Sensor-Based of Athlete Flexibility Measurement Tool

This study aims to develop a novel "sit and reach test" flexometer device utilizing a linear potentiometer sensor to quantitatively evaluate an individual's body flexibility. The innovation in this research lies in the utilization of electronic means to measure flexibility, replacing conventional manual methods. The device employs a linear potentiometer sensor that translates analog voltage into a digital value (ADC) and subsequently converts the digital data into distance measurements in centimeters (cm). The processed data, representing the distance measurements, is wirelessly transmitted to a personal computer (PC) for automatic capture and analysis. The design considerations prioritize modernity, practicality, effectiveness, and efficiency. The experimental outcomes demonstrate the efficacy of the developed tool. The readings from the linear potentiometer sensor exhibit a high linearity, indicated by an R2 value of 0.9999. The average error percentage is minimal, measuring at 0.17%. Moreover, the device allows for direct wireless transmission of data to a PC immediately after assessing an athlete's flexibility. This study not only introduces a novel electronic approach for assessing body flexibility but also validates its accuracy and efficiency through comprehensive testing and analysis.

Defect detection in additive manufacturing using image processing techniques

Additive manufacturing (AM) allows to produce parts layer by layer from an STL file. It is then possible through this technology to obtain customized geometries and complex shapes at a lower cost. However, these shapes pose problems of defect control, microstructure, residual stresses, and deformations in parts. This study aims to develop an efficient method allowing defect detection while printing pieces using Fused Deposition Modelling (FDM). The monitoring system contains a camera acquisition system for automatic image capture of filament layers deposited on the print bed. Various monitoring techniques have been simulated to achieve an optimal defect correction solution. Material excess and deficiency are detectable in the layer of actual printed parts. Defects are controlled and compared to original part obtained from Computer Aided Design (CAD). An app-designer application was created in this regard. It displays the image reference generated from the G-code, the layer image captured by the camera, and returns the error percentage in the printed layers. The developed method of surface calculation has shown its efficiency in detecting the lack and excess of material, which has an accuracy of 1.07%. This method allows users to stop and monitor printing to save cost, material, and time.

Visual Features for Improving Endoscopic Bleeding Detection Using Convolutional Neural Networks.

The presented paper investigates the problem of endoscopic bleeding detection in endoscopic videos in the form of a binary image classification task. A set of definitions of high-level visual features of endoscopic bleeding is introduced, which incorporates domain knowledge from the field. The high-level features are coupled with respective feature descriptors, enabling automatic capture of the features using image processing methods. Each of the proposed feature descriptors outputs a feature activation map in the form of a grayscale image. Acquired feature maps can be appended in a straightforward way to the original color channels of the input image and passed to the input of a convolutional neural network during the training and inference steps. An experimental evaluation is conducted to compare the classification ROC AUC of feature-extended convolutional neural network models with baseline models using regular color image inputs. The advantage of feature-extended models is demonstrated for the Resnet and VGG convolutional neural network architectures.

How far can the self be extended? Automatic attention capture is triggered not only by the self-face.

The preferential processing of self-related information is thought to be driven by its high level of familiarity. However, some behavioral studies have shown that people may exhibit a preference for initially unfamiliar stimuli that have been associated with themselves arbitrarily. One of the key questions that needs to be addressed concerns the role of early attention in the prioritization of newly acquired information associated with the self. Another question is whether both highly familiar as well as new information referring to a subjectively significant person (i.e. close-other) benefits from preferential attentional processing. We aimed to tackle both questions by investigating the neural mechanisms involved in processing extremely familiar stimuli, like one's own face or the face of a close-other, as well as stimuli (abstract shapes) that were newly linked to each person. We used a dot-probe paradigm that allowed us to investigate the early stages of attentional prioritization. Our analysis of the N2pc component unveiled that attention was automatically captured by the self-face, a shape associated with oneself, and the face of the close person. However, a shape associated with the close-other did not elicit the same attentional response, as the N2pc was absent. Thus, both the self-face and information referring to the extended self (self-assigned shape, close-other's face) benefit from preferential early and automatic attentional processing.

Study on the anticorrosion and antifouling performance of magnetically responsive self-healing polyurethane coatings

The dispersivity of self-healing microcapsules plays a critical role in the self-repairing properties of coatings. However, resolving the aggregation phenomenon of self-healing microcapsules has been challenging due to their high surface energy. In this paper, we successfully developed magnetically responsive self-healing microcapsule (MS-TO@CA) and polyurethane (PU) coatings with biomimetic interface wet-adhesion patterned microstructures and realized the adsorption, automatic capture and dispersion of MS-TO@CA based on different patterned microstructure sizes, microparticle sizes and contents. COMSOL software was used to simulate the magnetic field design and distribution, and the magnetization phenomenon of the microstructure was analyzed. The optimal process for the preparation of a new type of magnetically responsive self-healing PU coating was proposed. Compared with the control group, the corrosion time of the metal under the action of the coating was delayed by 816 h, and the antifouling performance was improved by approximately 30 times. This work has guiding significance for prolonging the service life of metals and improving metal anticorrosion and antifouling.

Retracted: Automatic Capture Processing Method of Basketball Shooting Trajectory Based on Background Elimination Technology

Retracted: Automatic Capture Processing Method of Basketball Shooting Trajectory Based on Background Elimination Technology

Use of photo-driven algal-bacterial aerobic granular sludge system to close carbon cycle in biological nutrients removal and sludge anaerobic digestion units in wastewater treatment plants

This study established a closed photo-driven algal-bacterial aerobic granular sludge (AGS) system coupling with an automatic biogas capture unit and with a working volume of 800 mL operated at a gas circulation rate of 0.75 L/min and liquid pH controlled at 8.0. This closed system was used to investigate the feasibility of algal-bacterial AGS to capture and fix carbon (C) from nutrients removal and anaerobic digestion (AD) processes. Results showed a high total C emission of 0.93 kg-CO2/kg-chemical oxygen demand (COD) under the external O2 supply condition. However, under the photo-driven mode with photosynthesis as the sole O2 supplier, the coexisting microalgae fixed all inorganic C from the nutrients removal process with a negative C emission of −0.25 kg-CO2/kg-COD and −0.33 kg-CO2/kg-COD under illumination at 24 klux and 36 klux, respectively. During this process more HCl was required to neutralize the increased pH under a stronger light intensity, which can be replaced by biogas (as an acid reagent) automatically introduced into the closed system as needed. Meanwhile, a CO2 removal rate of 40 mg-C/(g-mixed liquor volatile suspended solids (MLVSS)•d) and a biogas upgrading rate of 184 mL-biogas/(g-MLVSS•d) were achieved under 24 klux. It was estimated that the CO2 removal capacity can cover all CO2 produced from AD of the grown biomass in this system. In addition, photosynthetic O2 could simultaneously facilitate aerobic phosphorus uptake and ammonia oxidation at a low dissolved oxygen of 0.7–1.4 mg/L. Results from this work suggest that a properly designed and operated algal-bacterial AGS is promising for closing C cycle in wastewater treatment plants.