Scene Location Research Articles

Selective Brain Activations and Connectivities Related to the Storage and Recall of Human Object-Location, Reward-Location, and Word-Pair Episodic Memories.

Different cortical systems to the hippocampus were activated using fMRI during different types of episodic memory task. For object with scene location episodic memory, the activations were high in cortical systems involved in spatial processing, including the ventromedial visual and medial parahippocampal system. These activations for the medial parahippocampal system were higher in the right hemisphere. The activations in the face and object processing ventrolateral visual cortical stream regions FFC, PIT, V8 and TE2p were higher in the object-location in scene task than the reward-location task, and were higher in the right hemisphere. For reward-location in scene episodic memory, activations were also high in the ventromedial visual cortical spatial stream to the hippocampus, but were also selectively high in storage in key reward cortical regions (ventromedial prefrontal 10r, 10v, 10d; pregenual anterior cingulate d32, p24, p32, s32; and medial orbitofrontal cortex reward-related pOFC, 11l, OFC). For word-pair episodic memory, activations were lower in the ventromedial visual and medial parahippocampal spatial cortical stream, and were higher in language-related regions in Broca's area (44, 45, 47l), and were higher in the left hemisphere for these regions and for the many highly connected inferior frontal gyrus regions in the left hemisphere. Further, effective connectivity analyses during the episodic memory tasks showed that the direction of connectivity for these systems was from early visual cortical regions V2-V4 to the ventromedial visual cortical regions VMV1-3 and VVC for spatial scene processing; was from the pregenual anterior cingulate and orbitofrontal cortex reward systems to the hippocampal system; and was from the FFC/V8/PIT system to TE2p in the visual inferior temporal visual cortex, which has connectivity to lateral parahippocampal TF, which in turn has forward effective connectivity to the hippocampus.

Tobacco, e-cigarette and alcohol content in popular UK soap operas: a content analysis to explore changes in social norms and scene location

Background Exposure to tobacco and alcohol on-screen promotes use and despite regulations and policies to limit impact, these behaviours remain common. We report a longitudinal analysis of tobacco, e-cigarette and alcohol content in three popular UK television soap operas, to examine changing social norms between 2002 and 2022. Methods We used one-minute interval coding to measure content in programmes in two one-week periods in three years (2002, 2012 and 2022). Change in the probability of actual and implied use of tobacco, e-cigarette and alcohol over time was examined using logistic regression. Results We coded 2505 intervals from 78 episodes. Tobacco content occurred in 22% of episodes and significantly decreased from 2002 to 2022 (OR 0.15 95% CI 0.06–0.40). Tobacco use changed over time with decreasing use indoors and increasing use outdoors. No e-cigarette use was identified. Alcohol content was found in 88% of episodes and while it also significantly decreased over time (OR 0.78 95% CI 0.61–0.99) it featured in 20% of broadcast minutes in 2022. Alcohol use in homes increased over time. Conclusion While tobacco imagery is increasingly rare in these three UK soap operas, alcohol content has remained common. Tightening the UK Ofcom regulations would help to reduce young people’s exposure to these harmful behaviours and their potential influence on social norms now and in the future.

Massacres in the Arnsberg Forest (abstract)

In recent years, an historical reappraisal by one of us (MW) has been carried out on one of the worst crimes outside of prisons and concentration camps committed in Germany by the SS and Wehrmacht in the final months of the Second World War. In the Arnsberg Forest near Warstein and Meschede (southern Westphalia, western Germany), 208 forced labourers were massacred by the 'Division for Vengeance' (Division zur Vergeltung) of SS General Hans Kammler in March 1945. In the course of this joint project, archaeological research conducted by the Olpe Department of the LWL-Archäologie für Westfalen (Westphalian Archaeological Heritage Service) has pinpointed the exact location of the crime scenes and provided clues for the reconstruction of the course of events in the field (Warstein Langenbachtal, Suttrop, Meschede-Eversberg). It has also enabled the recovery of finds belonging to both the victims and perpetrators. We have also been able to locate and examine the two temporary cemeteries (Warstein Melkeplätzchen, Suttrop). This has provided information about the history of exhumation of the victims in 1964, as well as clarifying the whereabouts of the Soviet obelisk from the Melkeplätzchen, the provisional cemetery of the Langenbach victims located close to the crime scene. This work has also allowed a follow-up search in the temporary cemetery in Suttrop for seven victims who, according to the exhumation documents of 1964, are presumably still lying there. Between 2018 and 2021, almost all of aspects addressed above could be clarified with new insights gained. Our results were summarised at the EAC conference in Bonn 2023 (although not published in this theme). Several publications already report on them and they are listed below. In addition to several German language publications, there is also an extensive English language publication (Baales et al. 2022) where the essential aspects of the interdisciplinary project are presented. In addition, the historical context, research project and numerous images are presented on several websites listed below.

Does destination make a difference? Outcomes after a policy change affecting cutoff times for prehospital transport.

Facilitating primary triage and care at pediatric trauma centers (PTCs) can improve outcomes for children after trauma. However, scene location and regional emergency medical services regulations may result in initial evaluation occurring at nonpediatric facilities with later transportation to PTCs for definitive care. In this study, we assessed the results of a change in transport time cutoff from 30 to 45 minutes on pediatric patient outcomes. After institutional review board approval, the Pediatric Trauma Database at a level 1 PTC was queried for patients seen before (January 1, 2015, to December 31, 2017) and after (January 1, 2018, to December 31, 2020) the implementation of a policy increasing transport cutoff time from 30 to 45 minutes. Patient outcomes were compared by transport status and Injury Severity Score (ISS) using generalized linear regression analysis. A total of 505 patients were seen before policy changes, and 413 patients, after policy changes. Both groups had similar numbers of severely injured patients (ISS, ≥15; pre, 64 [13%]; post, 61 [15%]). Average transport time increased after change (pre, 20 minutes [95% confidence interval, 18-22 minutes]; post, 29 minutes [95% confidence interval, 26-33 minutes]; p = 0.0252), consistent with policy compliance. The proportion of transferred patients did not change after policy implementation ( p = 0.5856), and the complications among all patients with an ISS of ≥15 did not significantly decrease (pre, 75%; post, 65.6%). However, those patients with an ISS of ≥15 admitted directly from the scene had a lower frequency of complications after the policy changes (pre, 76%; post, 59%; p = 0.0319), and in the postperiod, transferred patients with an ISS of ≥15 had a higher complication rate than those admitted directly from the scene ( p < 0.0001). Direct scene admission to a PTC is associated with a lower complication profile for patients with higher ISS. Methods to ensure adherence to cutoff thresholds for emergency medical services transport may have a positive benefit on patient outcomes. Prognostic and Epidemiological; Level IV.

Refixation behavior in naturalistic viewing: Methods, mechanisms, and neural correlates.

When freely viewing a scene, the eyes often return to previously visited locations. By tracking eye movements and coregistering eye movements and EEG, such refixations are shown to have multiple roles: repairing insufficient encoding from precursor fixations, supporting ongoing viewing by resampling relevant locations prioritized by precursor fixations, and aiding the construction of memory representations. All these functions of refixation behavior are understood to be underpinned by three oculomotor and cognitive systems and their associated brain structures. First, immediate saccade planning prior to refixations involves attentional selection of candidate locations to revisit. This process is likely supported by the dorsal attentional network. Second, visual working memory, involved in maintaining task-related information, is likely supported by the visual cortex. Third, higher-order relevance of scene locations, which depends on general knowledge and understanding of scene meaning, is likely supported by the hippocampal memory system. Working together, these structures bring about viewing behavior that balances exploring previously unvisited areas of a scene with exploiting visited areas through refixations.

Where can offenders be found when they are not committing crimes? A cross-sectional survey of British police officers on non-crime locations

To date, the bulk of research on place and crime has concentrated on crime scene locations and home addresses of both victims and perpetrators. Beyond these locations, less is known about where offenders can be found during regular, non-criminal activity. These ‘non-crime locations’ provide information about offenders' lifestyles, their relationships with co-offenders, their preferred spots, and the places beyond the spatiotemporal boundaries of the crime where they may plan, manage, or deal with the aftermath of their behaviour. Yet there is a lack of systematic evidence, with the available research relying on local ethnographic designs, small samples or both, as access to such data was limited. This study utilises a novel approach by relying on the professional experience of intelligence police officers assigned to deal with serious and organised crime in the West Midlands County of the United Kingdom to determine the whereabouts of three types of non-crime locations: (a) hangout spots, (b) recruitment spots and (c) contraband hideout spots. Results suggest that offenders frequently visit gyms, specific restaurants, or private residences (not unlike locations where normative people hang out), and new members can be recruited anywhere. Drugs and firearms are strategically hidden near vulnerable persons. According to police officers, offenders are unconcerned by surveillance in these spots and choose venues not usually owned by criminal groups. Theoretical, methodological, and practical implications are discussed.

Conscious and unconscious memory and eye movements in context-guided visual search: A computational and experimental reassessment of Ramey, Yonelinas, and Henderson (2019)

Are eye movements unconsciously guided towards target locations in familiar scenes? In a recent eyetracking study, Ramey, Yonelinas, and Henderson (2019) measured eye-movement efficiency (scanpath ratio) and memory judgments when participants searched for targets in repeated and novel scenes. When trials judged new with high confidence were selected, scanpath ratio was lower for old scenes (misses) than for new scenes (correct rejections). In addition, familiarity as measured by recognition confidence did not significantly predict scanpath ratio. Ramey et al. attributed these results to unconscious learning guiding eye movements. In a re-assessment of Ramey et al.’s data, we show that their findings can be accounted for by a single-system computational model in which eye movements and memory judgments are driven by a common latent memory representation. In particular, (a) the scanpath ratio difference between high-confidence misses and correct rejections is a consequence of regression to the mean, while (b) the null correlation between familiarity and scanpath ratio, partly a natural consequence of the low reliability of the scanpath ratio measure, is also reproduced by the model. Two pre-registered experiments confirm a novel prediction of the alternative single-system model. This work offers a parsimonious account of Ramey et al.’s findings without recourse to unconscious guidance of eye movements.

Prehospital times and outcomes of patients transported using an ambulance trauma transport protocol: A data linkage analysis from New South Wales Australia

IntroductionPrehospital trauma systems are designed to ensure optimal survival from critical injuries by triaging and transporting such patients to the most appropriate hospital in a timely manner. ObjectivesWe sought to evaluate whether prehospital time and location (metropolitan versus non-metropolitan) were associated with 30-day mortality in a cohort of patients transported by road ambulance using a trauma transport protocol. MethodsData linkage analysis of routinely collected ambulance and hospital data across all public hospitals in New South Wales (NSW). The data linkage cohort included adult patients (age ≥ 16years) transported by NSW Ambulance, where a T1 Major Trauma Transport Protocol was documented by paramedic crews and transported by road to a public hospital emergency department in NSW for two years between January 2019 and December 2020. The outcomes of interest were prehospital times (response time, scene time and transport time) and 30-day mortality due to injury. Results9012 cases were identified who were transported to an emergency department with T1 protocol indication. Median prehospital transport times were longer in non-metropolitan road transports [n = 3,071, 98 min (71–126)] compared to metropolitan transports [n = 5,941, 65 min (53–80), p < 0.001]. There was no significant difference in 30-day mortality between the two groups (1.24% vs 1.65%, p = 0.13). In the subgroup of patients with abnormal vital signs, the only predictors of mortality were increasing age, presence of severe injury (OR 24.87, 95%CI 11.02, 56.15, p < 0.001), and arrival at a non-trauma facility (OR 3.01, 95%CI 1.26, 7.20, p < 0.05). Increasing transport times were not found to increase the odds of 30-day mortality. DiscussionIn the context of an inclusive trauma system and an established prehospital major trauma protocol, increasing prehospital transport times and scene location were not associated with increased mortality.

Bangsawan Dendam Laksamana' Backdrop: New Breath in Bangsawan Theater Performance

Bangsawan is an art form that combines elements of acting, singing, music and dance. Bangsawan is believed to have been brought to Malaya around the 1870s and adapted from Persian puppet shows in India. The nobility experienced a period of fame in the 1890s and began to experience its downfall in the early 1940s. Bangsawan to this day is still trying to compete with modern theater performances especially in Kuala Lumpur. 'Bangsawan Dendam Laksamana' was staged on 10-13 November 2017 and 2-11 November 2018 at the Kuala Lumpur City Hall Auditorium (DBKL), directed by Pat Ibrahim and published by Pat Ibrahim Project Entertainment. This study looks at the transformation made in the conventions of Bangsawan performances, especially looking at the innovations featured in the curtain background in this traditional theater. The curtain setting is a major form of design in the conventions of Bangsawan theater performances. Bangsawan uses seven types of curtain backgrounds that are used for each scene in his performance. The curtain background is the background of the scene location painted on a canvas measuring 20 feet wide and 15 feet high according to the suitability of the stage and requires a high cost and a long period of time to prepare it. 'Bangsawan Dendam Laksamana' composed the use of curtains from hand-painted fabric to a digital backdrop from an LED screen display. Nevertheless, the retention of the painting image on the curtain background is still done and improved by using digital painting on the LED. This has a modernizing impact in Bangsawan theater. This study discusses the extent to which reforms are made to elevate Bangsawan in trying to compete with contemporary theater today in Malaysia.

UAV Localization in Low-Altitude GNSS-Denied Environments Based on POI and Store Signage Text Matching in UAV Images

Localization is the most important basic information for unmanned aerial vehicles (UAV) during their missions. Currently, most UAVs use GNSS to calculate their own position. However, when faced with complex electromagnetic interference situations or multipath effects within cities, GNSS signals can be interfered with, resulting in reduced positioning accuracy or even complete unavailability. To avoid this situation, this paper proposes an autonomous UAV localization method for low-altitude urban scenarios based on POI and store signage text matching (LPS) in UAV images. The text information of the store signage is first extracted from the UAV images and then matched with the name of the POI data. Finally, the scene location of the UAV images is determined using multiple POIs jointly. Multiple corner points of the store signage in a single image are used as control points to the UAV position. As verified by real flight data, our method can achieve stable UAV autonomous localization with a positioning error of around 13 m without knowing the exact initial position of the UAV at take-off. The positioning effect is better than that of ORB-SLAM2 in long-distance flight, and the positioning error is not affected by text recognition accuracy and does not accumulate with flight time and distance. Combined with an inertial navigation system, it may be able to maintain high-accuracy positioning for UAVs for a long time and can be used as an alternative to GNSS in ultra-low-altitude urban environments.

Visual tracking in complex scenes: A location fusion mechanism based on the combination of multiple visual cognition flows

In recent years, deep learning has revolutionized computer vision and has been widely used for monitoring in diverse visual scenes. However, in terms of some aspects such as complexity and explainability, deep learning is not always preferable over traditional machine-learning methods. Traditional visual tracking approaches have shown certain advantages in terms of data collection efficiency, computing requirements, and power consumption and are generally easier to understand and explain than deep neural networks. At present, traditional feature-based techniques relying on correlation filtering (CF) have become common for understanding complex visual scenes. However, current CF algorithms use a single feature to describe the information of the target and locate it accordingly. They cannot fully express changeable target appearances in a complex scene, which can easily lead to inaccurate target locations in time-varying visual scenes. Moreover, owing to the complexity of surveillance scenes, monitoring algorithms can lose their target. The original template update strategy uses each frame with a fixed interval length as a new template, which may lead to unreliable feature extraction and low tracking accuracy. To overcome these issues, in this work, we introduce an original location fusion mechanism based on multiple visual cognition processing streams to achieve real-time and efficient visual monitoring in complex scenes. First, we propose a process for extracting multiple forms of visual cognitive information, and it is periodically used to extract multiple feature information flows of a target of interest. Subsequently, a cognitive information fusion process is employed to fuse the positioning results of different visual cognitive information flows to achieve high-quality visual monitoring and positioning. Finally, a novel feature template memory storage and retrieval strategy is adopted. When the location result is unreliable, the target is retrieved from memory to ensure robust and accurate tracking. In addition, we provide an extensive set of performance results showing that our proposed approach exhibits more robust performance at a lower computational cost compared with 36 state-of-the-art algorithms for visual tracking in complex scenes.

Visual attention during seeing for speaking in healthy aging.

As we age, we accumulate a wealth of information about the surrounding world. Evidence from visual search suggests that older adults retain intact knowledge for where objects tend to occur in everyday environments (semantic information) that allows them to successfully locate objects in scenes, but may overrely on semantic guidance. We investigated age differences in the allocation of attention to semantically informative and visually salient information in a task in which the eye movements of younger (N = 30, aged 18-24) and older (N = 30, aged 66-82) adults were tracked as they described real-world scenes. We measured the semantic information in scenes based on "meaning map" ratings from a norming sample of young and older adults, and image salience as graph-based visual saliency. Logistic mixed-effects modeling was used to determine whether, controlling for center bias, fixated scene locations differed in semantic informativeness and visual salience from locations that were not fixated, and whether these effects differed for young and older adults. Semantic informativeness predicted fixated locations well overall, as did image salience, although unique variance in the model was better explained by semantic informativeness than image salience. Older adults were less likely to fixate informative locations in scenes than young adults were, though the locations older adults' fixated were independently predicted well by informativeness. These results suggest young and older adults both use semantic information to guide attention in scenes and that older adults do not overrely on semantic information across the board. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

ROAD: The Road Event Awareness Dataset for Autonomous Driving.

Humans drive in a holistic fashion which entails, in particular, understanding dynamic road events and their evolution. Injecting these capabilities in autonomous vehicles can thus take situational awareness and decision making closer to human-level performance. To this purpose, we introduce the ROad event Awareness Dataset (ROAD) for Autonomous Driving, to our knowledge the first of its kind. ROAD is designed to test an autonomous vehicle's ability to detect road events, defined as triplets composed by an active agent, the action(s) it performs and the corresponding scene locations. ROAD comprises videos originally from the Oxford RobotCar Dataset, annotated with bounding boxes showing the location in the image plane of each road event. We benchmark various detection tasks, proposing as a baseline a new incremental algorithm for online road event awareness termed 3D-RetinaNet. We also report the performance on the ROAD tasks of Slowfast and YOLOv5 detectors, as well as that of the winners of the ICCV2021 ROAD challenge, which highlight the challenges faced by situation awareness in autonomous driving. ROAD is designed to allow scholars to investigate exciting tasks such as complex (road) activity detection, future event anticipation and continual learning. The dataset is available at https://github.com/gurkirt/road-dataset; the baseline can be found at https://github.com/gurkirt/3D-RetinaNet.

New Normal Creative Pop-Culture Tourism Promotion in Chiang Mai, Lamphun, and Lampang for Chinese Tourists

Purpose: The objectives of this study were 1) to study the behaviors of Chinese tourists who visit pop-culture destinations in Thailand’s Lanna tourism cluster, 2) To study and compare the decision-making factors of the Chinese tourists to visit Lanna tourism culture, and 3) to suggest provincial policies for Chinese tourists who visit pop-culture destinations in Thailand’s Lanna tourism cluster. Theoretical framework: This research figured out the demographic profiles of Chinese tourists who visited pop-culture destinations in Thailand’s Lanna tourism cluster. Then, the behaviors of Chinese tourists and the decision-making factors to visit pop-culture destinations in Thailand’s Lanna tourism cluster were further examined. Lastly, all data were combined to create the provincial policies for Chinese tourists who visited pop-culture destinations in Thailand’s Lanna tourism cluster. Design/methodology/approach: This research is a qualitative research methodology (QUAN). Total 412 samples of Chinese tourists who visited pop-culture destinations in Thailand’s Lanna tourism cluster were collected using verified questionnaires. The lists of questions investigated demographic profiles and behaviors during their trips of film tourism in Lanna tourism cluster. Full data were analyzed using frequency, percentage, mean, standard deviation, and hypothesis test via T-test and One-Way ANOVA. The data collecting process took several months from September-December 2021. Findings: The data revealed that most Chinese visitors who visited pop-culture destinations in Thailand’s Lanna tourism cluster were female, aged 31-35 years old, married, graduated with bachelor’s degree, and worked as private company employees with a monthly salary around 5,001-8,000 CNY. Friends &amp; relatives were the first choice of travel companion. The most inspiring travel decision was movie plots. Movie plots and performances did not influence Chinese visitors with difference marital statuses. Also, visitors with different education backgrounds were not influenced to visit by the locations of movie scenes, the movie trends, the cultural origins, the casts’ personalities, the presses and medias, and the driving trends. Place, social origin, personality, press, and push factor did not influence visitors with different personal incomes. Three provincial policies were recommended, including an administrative aspect, public relation aspect, and quality control aspect, to align with the 3rd National Tourism Development Plan of Thailand 2022-2026. Research, Practical &amp; Social implications: Without proper driving forces by stakeholders, the policies would not be healthy implemented. Consequently, the local government agencies should form a provincial pop-culture tourism committee to capably drive the policies. Originality/value: The behavioral studies of Chinese tourists who visited pop-culture destinations in Thailand’s Lanna tourism cluster were hardly found in academic arena. As well, all data were synthesized using previous literatures to combine with recent research results to create the provincial policies for Chinese tourists who visit pop-culture destinations in Thailand’s Lanna tourism cluster.

The Florida Violent Death Reporting System 2019 Data Findings Including a Qualitative Focus on Suicide Deaths of Young Black Males

The Florida Violent Death Reporting System (FLVDRS) includes violent death data abstracted from death certificates (DC), medical examiner (ME) reports, and law enforcement (LE) records. Abstractors also produce narratives from the ME and LE information. This manuscript focuses on the findings from the 2019 FLVDRS. Quantitative data were reviewed using SPSS and STATA and MAXQDA was used for the qualitative analysis. Results showed that most deaths were suicides (69.0%). Males comprised most of all the violent deaths (77.4%). Qualitative analysis focused on available narratives of Black males regardless of ethnicity, ages 10–24 years, who died of suicide (n = 19). The most prominent codes from the qualitative analysis included scene location, suicide method, engagement with LE, known history of psychiatric conditions, known history of self-injurious behavior, and recent life events. Subcodes related to the parent codes are also provided for further explanation of the findings. Future research will allow confirmation of findings and development of targeted intervention programs.

Is there human DNA on cats

The transfer, persistence, prevalence, and recovery of DNA (DNA-TPPR) can be highly relevant in forensic investigations to evaluate the presence and/or actions of a person of interest (POI). Whilst the DNA-TPPR-related research has increased significantly over the last decade, there is a lack of data on companion animals and their relationship to human DNA transfer. Given the commonality of cats and dogs in households around the world, companion animals as receptors and vectors for DNA transfer can be highly relevant in cases involving animals as victims of a criminal offense, or cases requiring activity level evaluations. Samples were collected from an external area on the right side of 20 cats to determine the prevalence and sources of human DNA on this area. Preliminary data shows that human DNA is present on household cats, its source is mainly from household inhabitants. Further studies are required to elucidate the means and level of transfer of human DNA to and from cats and other household animals. This knowledge can be relevant to sample targeting in specific case circumstances and/or when considering possible means of the presence of a person’s DNA at the crime scene location.

BlockCrime: Blockchain and Deep Learning-Based Collaborative Intelligence Framework to Detect Malicious Activities for Public Safety

Detecting malicious activity in advance has become increasingly important for public safety, economic stability, and national security. However, the disparity in living standards incites the minds of certain undesirable members of society to commit crimes, which may disrupt society’s stability and mental calm. Breakthroughs in deep learning (DL) make it feasible to address such challenges and construct a complete intelligent framework that automatically detects such malicious behaviors. Motivated by this, we propose a convolutional neural network (CNN)-based Xception model, i.e., BlockCrime, to detect crimes and improve public safety. Furthermore, we integrate blockchain technology to securely store the detected crime scene locations and alert the nearest law enforcement authorities. Due to the scarcity of the dataset, transfer learning has been preferred, in which a CNN-based Xception model is used. The redesigned Xception architecture is evaluated against various assessment measures, including accuracy, F1 score, precision, and recall, where it outperforms existing CNN architectures in terms of train accuracy, i.e., 96.57%.

VISEL: A visual and magnetic fusion‐based large‐scale indoor localization system with improved high‐precision semantic maps

Multisource fusion localization is a mainstream scheme for acquiring accurate locations in complex indoor scenes. To overcome the interference of indoor structures on radio and illumination variation on visual features, the semantic maps provide an effective way for multisource fusion localization. However, due to the lack of visual depth information, solutions of indoor semantic maps suffer from large semantic segmentation errors for similar objects, which leads to the unstable performance of localization systems. To overcome the issue in semantic and fusion localization, we develop a localization system to demonstrate the use of restudy semantic map and self-adapting fusion localization would achieve centimeter-level positioning accuracy, termed VISEL. VISEL uses the proposed spatial attention-aware semantic model to enhance the discrimination of semantic features for capturing accurate semantic maps. On the basis of high-precision semantic maps, VISEL completes an enhanced particle filter fusion localization module with adaptive reassign weight to different localization modules, which successfully improves accuracy through complementary advantages between different signals while overcoming the drawbacks of each signal and interference of complex environment. The extensive experimental results show that VISEL outperforms current state-of-the-art positioning systems and achieves an average positioning accuracy of 0.4 m. VISEL utilizes semantic maps with depth features and enhanced particle filter to reduce the fusion localization error by 38%, which suggests the high-precision semantic maps with depth features could provide a robust solution for the fusion localization system for indoor complex scenes.

A high-precision forest fire smoke detection approach based on ARGNet

The occurrence of forest fires can lead to ecological damage, property loss, and human casualties. Current forest fire smoke detection methods do not sufficiently consider the characteristics of smoke with high transparency and no clear edges and have low detection accuracy, which cannot meet the needs of complex aerial forest fire smoke detection tasks. In this paper, we propose Adjacent layer composite network based on a recursive feature pyramid with deconvolution and dilated convolution and global optimal nonmaximum suppression (ARGNet) for high-accuracy detection of forest fire smoke. First, the Adjacent layer composite network is proposed to enhance the extraction of smoke features with high transparency and no clear edges, and SoftPool in it is used to retain more feature information of smoke. Then, a recursive feature pyramid with deconvolution and dilated convolution (RDDFPN) is proposed to fuse shallow visual features and deep semantic information in the channel dimension to improve the accuracy of long-range aerial smoke detection. Finally, global optimal nonmaximum suppression (GO-NMS) sets the objective function to globally optimize the selection of anchor frames to adapt to the aerial photography of multiple smoke locations in forest fire scenes. The experimental results show that the ARGNet parametric number on the UAV-IoT platform is as low as 53.48 M, mAP reaches 79.03%, mAP50 reaches 90.26%, mAP75 reaches 82.35%, FPS reaches 122.5, and GFLOPs reaches 55.78. Compared with other mainstream methods, it has the advantages of real-time detection and high accuracy.

Review on Image Caption Generation

With the rapid development of Deep learning, AI along with Computer Vision and Natural Language processing Image caption has become an interesting and complex task. Image caption generation is the process of generating textual description of the given image and it is a challenging task because it consists of apprehension of objects. If the machine will be programmed to accurately describe an image or environment like human vision, it will be highly beneficial for robotic vision, business and many more. In order to generate an effective description of the image, the machine needs to detect, recognize objects as well as understand the scene type or location, object properties, their relationships and their interactions with each other. In this paper, we focus on advanced image captioning techniques such as CNN (Convolutional Neural Network)-LSTM(Long Short Term Memory) to generate meaningful captions. and the advantages and limitations of each method are discussed.