Single Scene Research Articles

MONOCULAR DEPTH PREDICTION IN PHOTOGRAMMETRIC APPLICATIONS

Abstract. Despite the recent success of learning-based monocular depth estimation algorithms and the release of large-scale datasets for training, the methods are limited to depth map prediction and still struggle to yield reliable results in the 3D space without additional scene cues. Indeed, although state-of-the-art approaches produce quality depth maps, they generally fail to recover the 3D structure of the scene robustly. This work explores supervised CNN architectures for monocular depth estimation and evaluates their potential in 3D reconstruction. Since most available datasets for training are not designed toward this goal and are limited to specific indoor scenarios, a new metric, large-scale synthetic benchmark (ArchDepth) is introduced that renders near real-world scenarios of outdoor scenes. A encoder-decoder architecture is used for training, and the generalization of the approach is evaluated via depth inference in unseen views in synthetic and real-world scenarios. The depth map predictions are also projected in the 3D space using a separate module. Results are qualitatively and quantitatively evaluated and compared with state-of-the-art algorithms for single image 3D scene recovery.

NARRATIVE DECORATIVE MOTIVES ON TORAJA TRADITIONAL CARVING

This study aims to explain the format and techniques of non-sacred narrative carving in Toraja. This research is classified as a qualitative research carried out in the field, namely in Tana Toraja and North Toraja. Data sources (1) artifacts in the form of non-decorative motifs of Toraja carving, both found in Toraja and elsewhere; (2) human data sources, namely Toraja sculptors and Toraja community leaders; and (3) documents that are documents containing descriptions, drawings or photographs of non-sacred decorative motifs of Toraja carving. Data collection techniques used were observation, interviews, and documentation. The collected data was analyzed with an interactive analysis model from Miles and Huberman. The results showed that: (1) The format of narrative carving art which is non-sacred in Toraja is (a) the format of a single scene narrative, (b) the panoramic narrative format, (c) the format of the narrative silmultanate. (2) The technique of depicting narrative carving art in Toraja, namely: (a) a realistic depiction technique, (b) the technique of presenting space there are two ways: presenting two-dimensional and three-dimensional space, and (c) the technique of composing elements (composition) which applied is symmetrical composition, asymmetric, rhythm, protrusion of object, and unity.

End-to-End Multi-Resolution 3D Capsule Network for People Action Detection

In this paper, we propose an end-to-end multi-resolution three-dimensional (3D) capsule network for detecting actions of multiple actors in a video scene. Unlike previous capsule, network-based action recognition does not specifically concern with the individual action of multiple actors in a single scene, our 3D capsule network takes advantage of multi-resolution technique to detect different actions of multiple actors that have different sizes, scales, and aspect ratios. Our 3D capsule network is built on top of 3D convolutional neural network (3DCNN) that extracts spatio-temporal features from video frames inside regions of interest generated by Faster RCNN object detection. We first apply our method to the problem of detecting illegal cheating activities in a classroom examination scene with multiple subjects involved. Second, we test our system on the publicly available and extensively studied UCF-101 dataset. We compare our method with several state-of-the-art 3DCNN-based methods, first the multi-resolution 3DCNN, the single-resolution 3D capsule network, and a combination of both these models. We show that models containing 3D capsule networks have a slight advantage over the conventional 3DCNN and multi-resolution 3DCNN. Our 3D capsule networks not only perform a classification of said actions but also generate videos of single actions. Our experimental results show that the use of multi-resolution pathways in the 3D capsule networks make the result even better. Such findings also hold even when we use pre-trained C3D (convolutional 3D) features to train these networks. We believe that the multiple resolutions capture lower-level features at different scales. At the same time, the 3D capsule layers combine these features in more complex ways than conventional convolutional models.

Multi-Temporal Landsat-8 Images for Retrieval and Broad Scale Mapping of Soil Copper Concentration Using Empirical Models

Mapping soil heavy metal concentration using machine learning models based on readily available satellite remote sensing images is highly desirable. Accurate mapping relies on appropriate data, feature extraction, and model selection. To this end, a data processing pipeline for soil copper (Cu) concentration estimation has been designed. First, instead of using single Landsat scenes, the utilization of multiple Landsat scenes of the same location over time is considered. Second, to generate a preferred feature set as input to a regression model, a number of feature extraction methods are motivated and compared. Third, to find a preferred regression model, a variety of approaches are implemented and compared for accuracy. In this research, 11 Landsat-8 images from 2013 to 2017 of Gulin County, Sichuan China, and 138 soil samples with lab-measured Cu concentrations collected from the area in 2015 are used. A variety a metrics under cross-validation are used for comparison. The results indicate that multi-temporal images increase accuracy compared to single Landsat images. The preferred feature extraction varies based on the regression model used; however, the best results are obtained using support vector regression and the original data. The final soil Cu map generated using the recommended data processing pipeline shows a consistent spatial pattern with a ground-truth land cover classification map. These results indicate that machine learning has the ability to perform large-scale soil heavy metal mapping from widely available satellite remote sensing images.

What Is Going On? An Analysis of the Interaction Order

Seeing sociology visually adds a sense of realness to the viewer compared to only reading sociological texts. In this paper, I aim to provide an example of how a single scene from a feature film can be utilized as a practical and meaningful means to analyze a social situation and to help students of sociology to grasp key features of Goffman’s theory of interaction order. More precisely, the main aims of the paper are 1) to illustrate Goffman’s theory of the interaction order by identifying acts of disruption and alignment in interaction through a film clip; and 2) to attempt to analyze, in a Goffmanian sense, what is really going on in the situational interaction. The scene is from the 2013 American movie August: Osage County and follows a dinner of immediate family in the wake of the funeral of the hostess’s late husband. The normative and civilized interaction of the meal is, however, jeopardized by the hostile and provocative mood of the hostess, as she repeatedly disrupts the interaction order with attempts to mock and/or uncover the hidden and vulnerable truths of the immediate members of her family, exemplifying her power status in the particular situation. The dinner guests, however, try to overlook and resist the provocation of the hostess and stick to their predetermined roles to save and sustain their idealized selves (their faces) and the interaction order (the faces of others), In doing so they, on the one hand, discard the uncomfortable truths acclaimed by the hostess and, on the other, explain the hostess’s provocative actions in terms of their claim that she is unwell and in need of medical attention. Thus, the attacked dinner guests in the scene align more alliance to the interaction order than to truth itself.

DGInet: Dynamic graph and interaction-aware convolutional network for vehicle trajectory prediction

This paper investigates vehicle trajectory prediction problems in real traffic scenarios by fully harnessing the spatio-temporal dependencies between multiple vehicles. The existing GCN-based trajectory predictions are often considered in a single traffic scene without time attributes, complete interaction information, dynamic graph-based model, etc. Time and interaction aware models are more challenging than the existing ones. Despite very well does the graph-based model describe the relationship between driving vehicles, the critical problem in the traffic scene is how to deeply explore the spatio-temporal characteristics of dynamic graphs. Therefore, a novel dynamic graph and interaction-aware neural network model called as DGInet is proposed by combining a semi-global graph mechanism and an M-product based graph convolutional network, which are built into novel dual-network architecture in the entire model. The DGInet is built by exploiting the dynamic interaction in depth between driving vehicles in urban traffic scenarios, and then realized by utilizing semi-global graph convolution operations on the input data cell to capture the basic spatial interaction features of the driving scene. Meanwhile, the dynamic graph is further extracted by a novel M-product approach, in which the embedding of the model is then established along with the embedding of the semi-global network to perform the final embedding. Extensive experiments have been conducted on the two public datasets, named NGSIM and Apollo respectively, to show that our approach outperforms the existing ones with better performance and less computing time. Besides the real-world Shenzhen traffic dataset, China, is also developed to verify the effectiveness of our approach.

Image Synthesis Pipeline for CNN-Based Sensing Systems.

The rapid development of machine learning technologies in recent years has led to the emergence of CNN-based sensors or ML-enabled smart sensor systems, which are intensively used in medical analytics, unmanned driving of cars, Earth sensing, etc. In practice, the accuracy of CNN-based sensors is highly dependent on the quality of the training datasets. The preparation of such datasets faces two fundamental challenges: data quantity and data quality. In this paper, we propose an approach aimed to solve both of these problems and investigate its efficiency. Our solution improves training datasets and validates it in several different applications: object classification and detection, depth buffer reconstruction, panoptic segmentation. We present a pipeline for image dataset augmentation by synthesis with computer graphics and generative neural networks approaches. Our solution is well-controlled and allows us to generate datasets in a reproducible manner with the desired distribution of features which is essential to conduct specific experiments in computer vision. We developed a content creation pipeline targeted to create realistic image sequences with highly variable content. Our technique allows rendering of a single 3D object or 3D scene in a variety of ways, including changing of geometry, materials and lighting. By using synthetic data in training, we have improved the accuracy of CNN-based sensors compared to using only real-life data.

Investigation of Potential Prevailing Wind Impact on Land Surface Temperature at Gas Flaring Sites in the Niger Delta, Nigeria.

This research examines the effects of South prevailing wind on Land Surface Temperature (LST) retrieved from Earth Observation (EO) Satellites at 11 gas flaring sites in Rivers State, Niger Delta region, Nigeria. 7 Landsat 5 Thematic Mapper (TM) and 18 Landsat 7 Enhanced Thematic Mapper Plus (ETM+) from 17/01/1986 to 08/03/2013 with < 5 % cloud contamination were considered. All sites are located within a single Landsat scene (Path 188, Row 057). The atmospherically corrected reflectance was used for the classification of 4 land cover (LC) types at each site. The emissivity (ε) for each site is estimated by using standard values for determined LC from Look Up Table (LUT). The surface-leaving radiance (Lλ) is computed from the atmospherically corrected thermal band 6 (High gain) and the emissivity (ε) values. The Planck equation was inverted using Landsat calibration constants to derive LST. Geospatial analysis of LST results using ArcGIS show 6 ranges of LST values for all sites. For both sensors LST retrieved for the flare stack sources are the highest values compared to other locations within the sites. Wind directions and wind speeds for Landsat data acquisitions dates and the South prevailing wind were applied to the LST for assessing their effects on it. The results show that for Eleme I and II, and Onne, the p-values results showed that no statistically significant relationships between δLST values in different directions (δLSTNE, δLSTNS and δLSTNW) existed. For Obigbo site, the wind direction (South) for data acquisition date combined with the South prevailing wind to generate a noticeable impact on the LST towards the North-East and the North-West directions. For Alua, Bonny, Chokocho, Rukpokwu, Umurolu and Sara sites, the p-value obtained is statistically significant for all the 3 (δLSTNE, δLSTNS and δLSTNW) relationships; therefore, producing a circle flare δLST footprint. For Umudioga site, only δLSTN versus δLSTW is statistically significant, causing a noticeable effect on the flare δLST in the North-West direction. Based on these results, it can be concluded that the volume and rate of burning gas, and the speed of the South wind at the time of satellite overpass are major factors that determine the influence of the South prevailing wind on the LST retrieved at the flaring sites in the Niger Delta.

Benchmarking the Robustness of Object Detection Based on Near-Real Military Scenes

According to the technical requirements of intelligent development of auxiliary combat system, we construct a visual intelligent test platform. A near-real military scene dataset based on physical rendering is built, which contains 11,000 remote sensing images collected by an analog camera taking pictures in different illumination, weather environment, camera shooting angle, and scene scale condition. Besides, we add a natural style transfer module for a single unmodeled military scene image’s multienvironment generation. We conduct experiments to evaluate the stability of several UAV remote sensing image object detection algorithms. Based on the quality and speed value of the tested algorithms, the adaptability scores in different environments are calculated. Furthermore, we propose a comprehensive evaluation index system of military remote object detection based on a hierarchical model. We envision that our comprehensive benchmark will play a role in the evaluation of algorithm capability for military object detection tasks and the improvement of training algorithm capability.

Multi-Agent Driven Resource Allocation and Interference Management for Deep Edge Networks

Sixth generation mobile networks (6G) may experience a huge evolution on vertical industry scenarios, where deep edge networks ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mathbf{D}\mathrm{E}\mathrm{N}\mathrm{s}$</tex-math></inline-formula> ) become an important network structure for the intelligent resource management of computing, caching and communications (3 C). In this paper, we propose two pervasive scenarios including single edge scene and multiple edge scenes for deep integration of wireless communications and computation depending on real-time adaptive collaboration. Specifically, in the single edge scenario, a novel deep reinforcement learning (DRL)-based framework is invoked for collaboratively optimizing the task scheduling, transmission power and CPU cycle frequency under metabolic channel conditions. Meanwhile, in order to alleviate interference in multiple edge scenarios, we propose a multi-agent aided deep deterministic policy gradient (MADDPG) algorithm to minimize total energy consumption and latency. Numerical experiments demonstrate these methods have substantial performance improvement in terms of saving system’s total overhead (i.e., energy consumption and delay) upon considering the task scheduling, transmission power, channel mutual interference and CPU cycle frequency against other conventional benchmarks.

HGA: Hierarchical Feature Extraction With Graph and Attention Mechanism for Linguistic Steganalysis

Linguistic steganalysis is an important topic in the field of information security and signal processing. In recent years, linguistic steganalysis have mainly utilized deep learning techniques and make great success. But suffer from the following major disadvantages. From the perspective of model structure, current methods only extract coarse features of the text, without focusing on the fine-grained representations. In terms of application, most of the studies only focus on single hidden scene and ignore the more realistic mixed hidden scenes which are more complex and realistic. These weaknesses limit the performance and the application of linguistic steganalysis in reality. In this paper, we propose a novel linguistic steganalysis method to overcome these weaknesses. This proposed method can extract distinguished text representation which fuses hierarchical features and perform excellently in sophisticated conditions. Firstly, we adapt gated graph neural networks as the coarse graph updater to update node representations on the graph level. Then we design a fine graph updater composed of the graph attention mechanism to focus on the highlighted nodes on the node-level. Moreover, we extract the most notable feature on the dimension-level of node by the graph channel attention module. Finally, the readout function is designed to fuse the hierarchical features and make the classification. The experimental results show that our method achieves the best results compared with the previous methods in both single hidden scene and mixed hidden scenes, which prove the effectiveness of the proposed method.

Feature Selection for Cross-Scene Hyperspectral Image Classification via Improved Ant Colony Optimization Algorithm

Hyperspectral images (HSIs) generally contain a large amount of spectral bands (features), and the redundant information in them will cause the Hughes phenomenon in the classification process. And feature extraction and feature selection are the two main existing methods to effectively reduce the redundancy of spectral information in the field of HSIs classification. Compared with feature extraction methods, feature selection methods can preserve most of the features of the original HSIs data without losing their valuable details. However, most existing feature selection methods based on single scene (domain) perform poorly in some scenes (domains) with insufficient labeled samples. Therefore, how to adopt an efficient feature selection method to select the optimal feature subsets of source scene and target scene and use the sample information of source scene to assist in the classification of target scene so as to improve the classification accuracy of images in the target scene as much as possible is still very challenging. In order to solve the above problem, this paper proposes a new cross-scene algorithm: Improved Ant Colony Optimization Algorithm-Based Cross-Scene Feature Selection Algorithm (IMACO-CSFS). In order to obtain more accurate feature subsets of the two scenes, IMACO-CSFS proposes a priority sorting-based ant colony strategy to make the subsequent search process focus on the global optimal solution (optimal feature subset) found in the previous iteration. In addition, in order to further accelerate the convergence speed of the global optimal solution, an ant colony strategy based on elite ants is proposed in IMACO-CSFS to more efficiently obtain the optimal feature subsets of the two scenes for training the classifier. Furthermore, this paper simultaneously considers overall classification accuracies of the optimal feature subsets for both scenes and dynamically adjusts their scale to ensure the consistency of the selected features between the two scenes, attenuating the effect of spectral shift and achieving the higher image classification accuracy in the target scene. Experimental results on three cross-scene HSI data pairs demonstrate that IMACO-CSFS is superior in cross-scene feature selection and cross-scene HSIs classification.

РАЗРАБОТКА ВИРТУАЛЬНОГО ПРОИЗВОДСТВА НА ПРИМЕРЕ УЗЛА РАЗДЕЛЕНИЯ ПЕНТАН-ГЕКСАНОВОЙ ФРАКЦИИ

Relevance The development of the educational process in educational institutions and enterprises requires appropriate improvement of computer training systems. For the training of highly professional personnel, the development of high-tech computer simulators using modern virtual reality technologies is required. AR and VR are a direction which popularity is growing from year to year. Today, almost every company is interested in applications using these technologies. From Oculus to MSQRD, from simple apps that project various toys in the room to IKEA view apps and more and so on. There are a huge variety of application options. But there is also not the most popular in comparison with them, but in fact a useful area – teaching a person new skills and simplifying his daily work. Here, as an example, we can cite simulators for doctors, pilots and even law enforcement agencies. Chemical and petrochemical enterprises are implementing these technologies as part of the digitalization of production. The main consumer is a direct employee of the production in gloves and a helmet, who is at the enterprise, at high-risk objects. The concept of «Industry 4.0» is becoming increasingly popular, the fourth industrial revolution is the unification of industry and digital technologies, leading to the creation of digital production facilities or smart factories and factories. In Europe, this concept is considered to be everything related to digital technologies in the enterprise in a general sense. In Russia, 4.0 is digital products that are somehow connected with hardware. First of all, of course, this is the industrial Internet of Things (IIoT), plus a direction related to video analytics, and also a direction called XR (AR + VR). The main task of IIoT is to increase the level of automation in production, reduce the influence of the human factor on the process of managing non-critical technological processes, and reduce the cost of equipment operation. Virtual reality is a model of the surrounding world, artificially created with the help of technical means and presented in digital form. The created effects are projected onto the human consciousness and allow you to experience sensations that are as close to real as possible. The analogues that exist today, developed by large design institutes and companies, have limited access and high cost. The virtual petrochemical production developed within the framework of this project should be a software product containing a production site as close as possible to the real one: technological equipment – columns, heat exchangers, tanks, facilities – pumps, compressors, pipelines, control valves, valves, sensors and actuators. The user, wearing virtual reality glasses and picking up controllers, will have the opportunity to take a tour of the developed production site. The visual series should almost completely correspond to reality, but some details that do not affect the learning process can be simplified so as not to spend a lot of time reproducing them. The use of sound effects that simulate the real sounds of this production will complement the effect of immersion in a virtual reality environment. Aim of research Transfer of the production process of the pentane-hexane fraction separation unit to a virtual reality environment. Methods To create 3d models of the necessary technological equipment, machines, devices, etc., the Autodesk 3ds Max 2015 program was used. Unreal Engine 4 was used to collect the created models into a single scene, as well as to write code for interacting with the virtual reality environment. Results This article discusses the creation of a model of the production process of the pentanehexane fraction separation unit in a virtual reality environment in order to monitor the progress of the technological process and control it through the objects of this production.

Unsupervised Single-Scene Semantic Segmentation for Earth Observation

Earth observation data has huge potential to enrich our knowledge about our planet. An important step in many Earth observation tasks is semantic segmentation. Generally, a large number of pixelwise labeled images are required to train deep models for supervised semantic segmentation. On the contrary, strong inter-sensor and geographic variations impede the availability of annotated training data in Earth observation. In practice, most Earth observation tasks use only the target scene without assuming availability of any additional scene, labeled or unlabeled. Keeping in mind such constraints, we propose a semantic segmentation method that learns to segment from a single scene, without using any annotation. Earth observation scenes are generally larger than those encountered in typical computer vision datasets. Exploiting this, the proposed method samples smaller unlabeled patches from the scene. For each patch an alternate view is generated by simple transformations, e.g., addition of noise. Both views are then processed through a two-stream network and weights are iteratively refined using deep clustering, spatial consistency, and contrastive learning in the pixel space. The proposed model automatically segregates the major classes present in the scene and produces the segmentation map. Extensive experiments on four Earth observation datasets collected by different sensors show the effectiveness of the proposed method. Implementation is available at https://gitlab.lrz.de/ai4eo/cd/-/tree/main/unsupContrastiveSemanticSeg.

Apple image fusion algorithm based on binocular acquisition system

To solve the problem that single natural scene image acquisition information in an orchard cannot meet the requirements of accurate fruit recognition and target positioning, a new apple image fusion algorithm based on the binocular acquisition system named New Non-sub-sampled Contourlet Transform algorithm is proposed to obtain a high-quality fusion image. The binocular acquisition system is constructed with the Time of Flight industrial camera and the colour camera. In order to achieve a better fusion effect, a parameter optimisation algorithm based on an Artificial Bee Colony algorithm with discard strategy for brightness saliency function is proposed to optimise the low-frequency component parameters of the new fusion coefficient rule. The experiments were taken on series of apple images under three sunlight conditions in the orchard. The experimental results show that the six evaluation indicators obtained by the new algorithm achieve the expected fusion image effect under three different types of sunlight conditions.

Distance Measurement in Industrial Scenarios

3D reconstruction technology is one of the important technologies of computer vision. Compared with 2D data, 3D space contains more abundant information, including location information, local / global features and so on. It is of great significance to solve the problem of target distance measurement in a single image scene. It is difficult for neural network to predict hole size without image scale information. In this paper, we focus on the measurement of objects with little change in height in the image. We use the relationship between camera parameters and regional features and the internal relationship between regional features to solve the problems of three-dimensional parameter reconstruction and monocular image distance measurement and use the standard cross entropy loss to optimize the transformer model. This model has achieved good results on the sampled data set.

Spectral Analysis With a Piecewise Separation Occlusion Model for Image-Based Rendering

Occlusion is ubiquitous in the real world, and neglecting it will certainly lead to quality degeneration in a reconstructed light field (LF). In this paper, we derive a piecewise separation occlusion model (PSOM) for LF reconstruction that is optimized for occlusion in the continuous Fourier and integral domains. The PSOM is first obtained by determining the definite integral intervals between the occlusion and nonocclusion boundaries in the spatial domain along the camera direction. We then use these definite integral intervals to perform spectral analysis of the plenoptic function (POF) to derive the occluded and nonoccluded plenoptic spectra. Finally, we show that using the two sampling rates derived from the occluded and nonoccluded spectra is more effective than using a uniform sampling rate for reconstructing a single scene unless the scene is free of occlusion. Through numerical experiments on simulated signals and real-world inward-looking and outward-looking scenes, we demonstrate the effectiveness of our proposed approach.

Predicting and Manipulating Cone Responses to Naturalistic Inputs.

Primates explore their visual environment by making frequent saccades, discrete and ballistic eye movements that direct the fovea to specific regions of interest. Saccades produce large and rapid changes in input. The magnitude of these changes and the limited signaling range of visual neurons mean that effective encoding requires rapid adaptation. Here, we explore how macaque cone photoreceptors maintain sensitivity under these conditions. Adaptation makes cone responses to naturalistic stimuli highly nonlinear and dependent on stimulus history. Such responses cannot be explained by linear or linear-nonlinear models but are well explained by a biophysical model of phototransduction based on well-established biochemical interactions. The resulting model can predict cone responses to a broad range of stimuli and enables the design of stimuli that elicit specific (e.g., linear) cone photocurrents. These advances will provide a foundation for investigating the contributions of cone phototransduction and post-transduction processing to visual function.SIGNIFICANCE STATEMENT We know a great deal about adaptational mechanisms that adjust sensitivity to slow changes in visual inputs such as the rising or setting sun. We know much less about the rapid adaptational mechanisms that are essential for maintaining sensitivity as gaze shifts around a single visual scene. We characterize how phototransduction in cone photoreceptors adapts to rapid changes in input similar to those encountered during natural vision. We incorporate these measurements into a quantitative model that can predict cone responses across a broad range of stimuli. This model not only shows how cone phototransduction aids the encoding of natural inputs but also provides a tool to identify the role of the cone responses in shaping those of downstream visual neurons.

Traffic incident detection based on a global trajectory spatiotemporal map

Traffic incidents endanger the smooth running of vehicles. Congestion caused by traffic incidents has caused a waste of time and fuel and seriously affected transportation efficiency. At present, most methods use manual judgment or image features to detect traffic incidents, but these methods lack timeliness, leading to secondary incidents. For dangerous road sections such as ramp-free and long downhills, this paper proposes an algorithm to quickly detect traffic incidents based on a spatiotemporal map of vehicle trajectories. First, a vehicle dataset from the monitoring perspective is constructed, and an improved YOLOv4 detection algorithm is used to detect images organized as batches. Based on the detection result, the multi-object tracking method of vehicle speed prediction in key frames is used to obtain the vehicle trajectory. Then according to the vehicle trajectory obtained in a single scene, the vehicle trajectory is reidentified and associated in the continuous monitoring scene to construct a long-distance vehicle trajectory spatiotemporal map. Finally, according to the distribution and generation status of the trajectory in the spatiotemporal map, traffic incidents such as vehicle parking, vehicle speeding, and vehicle congestion are analyzed. Experimental results show that the proposed method greatly increases the speed of vehicle detection and tracking and obtains high mAP, MOTA, and MOTP indicators. The global spatiotemporal map constructed by trajectory reidentification can achieve high detection rates for traffic incidents, reduce the average elapsed time, and avoid the problems of the inaccuracy of analyzing image features.

Single object tracking algorithm in occlusion scene based on improved ECO

To alleviate the occlusion problem in a single object tracking scene, this paper proposes an ECO-MHDU object tracking algorithm with a more powerful anti-occlusion performance based on the ECO tracker. The algorithm first uses the pre-trained MobileNetV3 lightweight backbone network on the ImageNet dataset to replace the ResNet network in the ECO to increase the speed of the algorithm to obtain the shallow and deep feature information of the image, while effectively using the attention mechanism in the MobileNetV3 network to strengthen the algorithm’s ability to extract target features; secondly, use the DropBlock operation on the acquired feature map to generate a random continuous mask on the feature map channel to improve the algorithm’s learning of the global robust spatial structure information; finally, a confidence update strategy is introduced into the GMM sample generation space. To improve the quality of training samples, unreliable tracking states such as confidence detection and occlusion are designed to avoid updating the sample space with damaging information. Compared with the ECO algorithm, the ECO-MHDU algorithm proposed in this paper has a success rate of 68.0% on the occlusion attributes of the OTB100 dataset, which is 2.3% higher than the ECO algorithm, and the ECO-MHDU algorithm also showed the best performance on the entire dataset sequence, with a success rate of 69.3%.