Object Identification Algorithm Research Articles

Global and Regional Characteristics of Radially Outward Propagating Tropical Cyclone Diurnal Pulses

AbstractThe radially outward propagating, cloud top cooling, diurnal pulse (DP) is a prominent feature in tropical cyclones (TCs) that has important implications for changes in TC structure and intensity. By using an objective identification algorithm, this study characterizes DPs over various ocean basins and examines their environmental conditions and convective structures. DPs occur on 52% of TC days globally and the occurrence frequency exhibits significant regional variability. The Northwest Pacific (NWP) has the highest DP frequency (60%) and shares the largest fraction of DPs worldwide (34%). The median duration and propagation distance of DPs are 12–15 hr and 500–600 km, respectively. Although the mean propagation speed of DPs is 11–13 m s−1, persistent DPs (lasting >15 hr) mostly propagate at speeds similar to internal inertial gravity waves (5–10 m s−1). Additionally, the longer the pulse duration, the stronger the pulse amplitude. Further, most DPs initiate in the inner core overnight, in phase with inner‐core deep convection. Inner‐core cold clouds, precipitation, and lightning are all markedly enhanced on DP days compared to non‐DP days. Interestingly, the DP signal significantly weakens and becomes slower while propagating through the 200–400‐km annulus during 09–12 local time (LT). Finally, DPs are more likely to occur over warm sea surface temperatures (SSTs), in low shear, and with a moist mid‐to‐upper troposphere. SST plays an important role in DP development over all basins, while shear and humidity are less important in the Northeast Pacific (NEP) and North Atlantic (NA) basins.

Drishti : Real-Time Object Recognition for the Visually Impaired

In 2017, the World Health Organization (WHO) reported that nearly 284 million individuals worldwide experienced some degree of visual impairment, with approximately 39 million individuals suffering from total blindness. People with visual impairments often rely on assistance from others or use canes to move around and identify obstacles. Our proposed system aims to aid the visually impaired by identifying and classifying common objects in real-time, as well as recognizing text from various sources such as documents and signs. This system provides voice feedback to enhance understanding and navigation, and utilizes depth estimation algorithms to determine a safe distance between objects and individuals, promoting self-sufficiency and reducing dependence on others. We employ the COCO image dataset, which contains everyday objects and people, and utilize the Mobilenet SSD algorithm for real-time object identification. To enable real-time Optical Character Recognition (OCR) Text-To-Speech functionality, we employ advanced technologies such as OpenCV, Python, and Tesseract for text detection and recognition, and the Pyttsx3 library for converting recognized text into audible speech. Our proposed system is dependable, affordable, realistic, and feasible.

Robustness Analysis of Traffic Sign Recognization based on ResNet

Autonomous driving has always been an important research topic and application task of artificial intelligence, which has attracted the attention of a large number of researchers. As an important component of the environmental perception module in autonomous driving tasks, traffic sign recognition can help drivers understand road information in a timely manner and avoid potentially dangerous driving operations. To this end, accurate recognition of traffic signs is crucial from both a strategic and a practical point of view. The early recognition technology of traffic signs is mainly based on the detection of color and shape, whose recognition accuracy is limited due to the fading and deformation of traffic signs. Numerous researchers have successfully used these deep learning-based object identification algorithms for traffic sign detection and recognition thanks to the development of the Faster R-CNN and YOLO series algorithms. However, we argue that few efforts focus on the recognition performance of the model in different scenarios, which is especially important in the process of autonomous driving. Based on this observation, this paper first constructs the ResNet model. Part of the image recognition accuracy predicted by the model reached 99%. Afterward, the robustness of the model is explored by simulating complex scenes by changing illumination and noise, and it is proved that the model has good generalization ability and practical application ability.

Pothole and Wet Surface Detection Using Pretrained Models and ML Techniques

Abstract: Roads contribute significantly to the economy and serve as a transportation platform. Road potholes are a key source of worry in transportation infrastructure. The purpose of this research is to develop an Artificial Intelligence (AI) model for identifying potholes on asphalt pavement surfaces. Image processing techniques from pretrained models such as efficientnet, resnet50, mobilenet and ML models such as random forest, decision tree, SVC, SVM. Several studies have advocated employing computer vision techniques, including as image processing and object identification algorithms, to automate pothole detection. It is important to digitize the pothole identification process with acceptable accuracy and speed, as well as to deploy the procedure conveniently and affordably. Initially, a smartphone placed on the automobile windscreen captures many photographs of potholes. Later, by downloading pothole photographs from the internet, we expanded the amount and variety of our collection (2400 images with over 900 potholes). Second, to locate potholes in road photos, several object detection methods are used. To compare pothole detection performance, real-time Deep Learning algorithms in various setups are employed. Similarly Wet pavement decreases surface friction dramatically, increasing the likelihood of an accident. As a result, timely understanding of road surface condition is essential for safe driving. This research proposes a unique machine learning model pipeline for detecting pavement moisture based on live photos of highway scenes acquired via accessible to the public traffic cameras. We refined existing state-of-the-art feature extraction baseline models to capture background instance targets, such as pavement, sky, and vegetation, which are frequent in highway scenes, to simplify the learning job.

Distinct characteristics of western Pacific atmospheric rivers affecting Southeast Asia

The dynamic characteristics of atmospheric rivers (ARs) have been researched over the western North Pacific and East Asia due to their close linkage to disastrous precipitation extremes, while very little attention has been paid to the AR features from the western Pacific to Southeast Asia. This study aims to quantify the climatology, long-term trends and variability of different AR properties from the western Pacific to Southeast Asia using an objective identification algorithm, the ERA5 reanalysis dataset and the APHRODITE precipitation dataset for the period 1951-2015. The results indicate a belt of frequent AR activities from the western Pacific to the Andaman Sea during the boreal autumn-winter season. The long-term trend analyses show a significantly increasing trend in AR frequency and an eastward shift of AR plumes. These dynamic changes contribute to the increasing trend of extreme precipitation amounts in the coastal areas surrounding the South China Sea. The intraseasonal variability of the AR associated with the Madden-Julian oscillation (MJO) shows a pronounced enhancement of AR activity in the MJO phase-2 to phase-3 due to the steeper gradient of low-level geopotential height between the Northwestern Pacific and the tropical Indian Ocean. The modulation is partly explained by the enhanced MJO convection and the adiabatic heating in the vicinity of the trough of the 200-500 hPa geopotential thickness of the region. This study shows that ARs are important mechanisms behind the climatology, trends and variability of the regional precipitation in Southeast Asia. This study implies that more attention is required toward the dynamics of these tropical weather systems, particularly for their interactions with other synoptic processes and their response to future climate warming.

Real-Time Image Processing Using Deep Learning With Opencv And Python

The observation of laptop imaginative and prescient aids in the improvement of techniques for figuring out presentations and pictures. It contains a variety of functions, including picture recognition, object identification and image production among others. Face recognition, vehicle recognition, online photos, and safety systems all employ object detection. The goal is to identify things using the You Only Look Once (YOLO) technique. When compared to previous object identification algorithms, our method focuses on a few key areas. Unlike other algorithms, YOLO scans the whole photograph through estimating bounding containers the use of convolutional networks and sophistication possibilities for those containers. This permits YOLO to understand an photograph extra fast than different algorithms together with convolutional neural networks and speedy convolutional neural network. By using dependencies like OpenCV, we can identify each object in an image based on the region object in a distinct rectangular box, identify every item and assign its tag to the item the use of those strategies and algorithms primarily based totally on deep learning, which is likewise primarily based totally on system learning. It moreover consists of the nuances of every item-marking strategy.

Pothole Detection and Management System using AI

Potholes put pedestrian and vehicular safety in peril, creating a traffic hazard. In the majority of developing nations, it ranks among the main factors in traffic accidents that result in the loss of life and property. In order to identify potholes quickly and accurately, this project aims to assess the performance of cutting-edge neural network techniques like YOLO and Faster R-CNN with VGG16 and ResNet-18 architectures. A new YOLOv2 architecture is also proposed to handle the "pothole" and "regular road" class imbalance issue, and its performance is evaluated in comparison to other object identification algorithms using accuracy, recall, intersection over union, and frames processed per second (FPS).

Innovative Region Convolutional Neural Network Algorithm for Object Identification

Object identification is a part of the field of computer science, namely, image processing, whose research continues to innovate. Object identification describes an object based on the main characteristics of the object. Many research innovations related to object identification have been carried out to obtain optimal identification results. The convolutional neural network (CNN) is one of the algorithms that is widely used by researchers in the field of object identification or object recognition in digital images. The purpose of this study was to analyze the development of object identification in the search for the best algorithm in terms of the speed and efficiency of identification. The article data used were obtained from several sources, namely, Dimensions AI, Science Direct, and Google Scholar. The database search results obtained 1041 articles in the form of publications from 2010–2021. Through a systematic literature review based on the articles obtained, 32 articles were selected. The evaluation of the articles was carried out in the form of article data visualization, object identification algorithm development, and the research objects used. CNN’s research innovation is growing rapidly, with improvements being made to the identification techniques in its algorithmic architecture. The use of the CNN algorithm in the identification of image objects, starting with the region CNN technique, is improved with Fast R-CNN, Faster-CNN, and Mask R-CNN. The object of research has developed from facial recognition and the identification of moving images to the introduction of ancient manuscripts that are useful for the development of history and tourism. The successful identification of ancient scripted texts will greatly assist the availability of such manuscripts in a digital format, which allows for further multidisciplinary research. The availability of ancient manuscripts in a digital format also helps the government to preserve culture and increase people’s understanding of the culture they have.

Single-Stage Underwater Target Detection Based on Feature Anchor Frame Double Optimization Network.

Objective: The shallow underwater environment is complex, with problems of color shift, uneven illumination, blurring, and distortion in the imaging process. These scenes are very unfavorable for the reasoning of the detection network. Additionally, typical object identification algorithms struggle to maintain high resilience in underwater environments due to picture domain offset, making underwater object detection problematic. Methods: This paper proposes a single-stage detection method with the double enhancement of anchor boxes and features. The feature context relevance is improved by proposing a composite-connected backbone network. The receptive field enhancement module is introduced to enhance the multi-scale detection capability. Finally, a prediction refinement strategy is proposed, which refines the anchor frame and features through two regressions, solves the problem of feature anchor frame misalignment, and improves the detection performance of the single-stage underwater algorithm. Results: We achieved an effect of 80.2 mAP on the Labeled Fish in the Wild dataset, which saves some computational resources and time while still improving accuracy. On the original basis, UWNet can achieve 2.1 AP accuracy improvement due to the powerful feature extraction function and the critical role of multi-scale functional modules. At an input resolution of 300 × 300, UWNet can provide an accuracy of 32.4 AP. When choosing the number of prediction layers, the accuracy of the four and six prediction layer structures is compared. The experiments show that on the Labeled Fish in the Wild dataset, the six prediction layers are better than the four. Conclusion: The single-stage underwater detection model UWNet proposed in this research has a double anchor frame and feature optimization. By adding three functional modules, the underwater detection of the single-stage detector is enhanced to address the issue that it is simple to miss detection while detecting small underwater targets.

Unique Radar Observations of Large Raindrops in Tropical Warm Rain during PISTON

Abstract Isolated warm-rain cells are an important feature over the tropical oceans. Although warm rain is typically associated with relatively small raindrops, large raindrops (>4.5 mm in diameter) have been observed in some cases. Previous studies have examined warm rain cells with large drops on a case-study basis, but they have yet to be investigated in a broader, statistical sense. During the recent Propagation of Intraseasonal Oscillations (PISTON) field campaign, a C-band polarimetric radar routinely measured extreme values of differential reflectivity in small, isolated convection, indicating the presence of large drops. Using an objective feature identification and tracking algorithm, this study offers new insights to the structure and frequency of cells containing large drops. Cells with high differential reflectivity (>3.5 dB) were present in 24% of all radar scans. The cells were typically small (8-km2 mean area), short lived (usually <10 min), and shallow (3.7-km mean height). High differential reflectivity was more often found on the upwind side of the cells, suggesting a size sorting mechanism was operating establishing a low concentration of large drops on the upwind side. Differential reflectivity also tended to increase at lower altitudes, which is hypothesized to be due to continued drop growth and increasing temperature (increasing the dielectric constant of water). Rapid vertical cross-section radar scans, as well as transects made by a Learjet aircraft with onboard particle probes, are also used to analyze these cells, and support the conclusions drawn from statistical analysis.

Combination of Deep Cross-Stage Partial Network and Spatial Pyramid Pooling for Automatic Hand Detection

The human hand is involved in many computer vision tasks, such as hand posture estimation, hand movement identification, human activity analysis, and other similar tasks, in which hand detection is an important preprocessing step. It is still difficult to correctly recognize some hands in a cluttered environment because of the complex display variations of agile human hands and the fact that they have a wide range of motion. In this study, we provide a brief assessment of CNN-based object identification algorithms, specifically Densenet Yolo V2, Densenet Yolo V2 CSP, Densenet Yolo V2 CSP SPP, Resnet 50 Yolo V2, Resnet 50 CSP, Resnet 50 CSP SPP, Yolo V4 SPP, Yolo V4 CSP SPP, and Yolo V5. The advantages of CSP and SPP are thoroughly examined and described in detail in each algorithm. We show in our experiments that Yolo V4 CSP SPP provides the best level of precision available. The experimental results show that the CSP and SPP layers help improve the accuracy of CNN model testing performance. Our model leverages the advantages of CSP and SPP. Our proposed method Yolo V4 CSP SPP outperformed previous research results by an average of 8.88%, with an improvement from 87.6% to 96.48%.

SHOPPING MALL PREMISES SCADA-MICROCLIMATE CONTROL SYSTEM DEVELOPMENT

The article is devoted to the shopping malls technological processes of heating, air conditioning and ventilation control studying. In order to develop a shopping mall microclimate control automated system temperature, humidity and ventilation control laws were analysed. Literature review shows that current microclimate control systems are not energy efficient and the usage of supervisory control and data acquisition systems can provide information of the current system state and decrease energy consumption. It is stated that using SCADA-system helps to reduce equipment operating costs by collecting and processing real-time information and implementing new energy-efficient control laws. To achieve this goal, the existing microclimate control systems were analysed. It was determined that these systems can be improved by using the SCADA system. Current implementations of shopping malls control parameters are investigated. Based on investigated control parameters the requirements of SCADA-systems construction were received. It was shown that the developed control system simplifies shopping malls microclimate control system construction, control object identification and energy-efficient control algorithms implementation. An overview of temperature and humidity processes in the shopping malls control algorithms was made. The list of input-output parameters of ventilation system control has been made. Using these parameters, the structural scheme of connection was built. The equipment of an automated control system is described. Operator visualization for microclimate parameters tracking have been developed. Visualizations of the authorization, room selection and setting control settings have been designed. The practical significance of the study lies in the hardware and software system control development.

A Climatology of Narrow Cold‐Frontal Rainbands in Southern California

AbstractNarrow cold‐frontal rainbands (NCFRs) produce short‐duration, high‐intensity precipitation associated with flash flooding and debris flows in Southern California, including the extremely destructive 2018 Montecito event. This manuscript presents the first known catalog of NCFR events in this region during winter seasons 1995–2020. A manual approach was used to identify NCFRs in radar imagery, supported by rain gauges and atmospheric reanalysis products. In total, 94 NCFRs were observed, with an average of three landfalling events occurring per water year. Propagation and impacts of these NCFRs were also investigated. Composite analyses of synoptic conditions reveal similarities across events; however, individual NCFRs exist on a spectrum of atmospheric conditions, precipitation intensity, and associated hazards. This regional record serves as a control dataset for the development of an objective identification algorithm and supports the evaluation and improvement of forecasting procedures. Methods established here are transferrable to other mid‐latitude regions with limited radar coverage.

A Drone System with an Object Identification Algorithm for Tracking Dengue Disease

In recent decades, it has been shown that epidemi-ological surveillance is one of the most valuable tool that public health has, since it allows us to have an overview of the population general health, thus allowing to anticipate outbreaks of epidemics by helping in timely interventions. Currently there is an increase in cases of dengue disease in several regions of Peru. Therefore, to control this outbreak and to help population centers and human settlements that are far from the city this work puts forward a drone system with an object recognition algorithm. Drones are very efficient in terms of surveillance, allowing easy access to places that are difficult for humans. In this way, drones can carry out the field work that is required in epidemiological surveillance, carrying out photography or video work in real time, and thus identifying infectious foci of diverse diseases. In this work, an object detection algorithm that uses convolutional neural networks and a stable detection model is designed, this allows the detection of water reservoirs that are possible infectious sources of dengue. In addition the efficiency of the algorithm is evaluated through the statistical curves of precision and sensitivity that result of the training of the neural network. To validate the efficiency obtained, the model was applied to test images related to dengue, achieving an efficiency of 99.2%.

Mesoscale Eddy-Induced Ocean Dynamic and Thermodynamic Anomalies in the North Pacific

Oceanic mesoscale eddies are associated with large thermodynamic anomalies, yet so far they are most commonly studied in terms of surface temperature and in the sense of composite mean. Here we employ an objective eddy identification and tracking algorithm together with a novel matching and filling procedure to more thoroughly examine eddy-induced thermodynamic anomalies in the North Pacific, their relationship with eddy amplitude (SSH), and the percentage of variability they explain on various timescales from submonthly to interannual. The thermodynamic anomalies are investigated in terms of sea surface temperature (SST), isothermal layer depth (ITD), and upper ocean heat content (HCT). Most eddies are weak in amplitude and are associated with small thermodynamic anomalies. In the sense of composite mean, anticyclonic eddies are generally warm eddies with deeper isothermal layer and larger heat content, and the reverse is true for cyclonic eddies. A small fraction of eddies, most probably subsurface eddies, exhibits the opposite polarities. Linear relationships with eddy amplitude are found for each of the thermodynamic parameters but with different level of scatter and seasonality. HCT-amplitude relation scatters the least and has the smallest seasonal difference, ITD-amplitude relation has the largest scatter and seasonality, while SST-amplitude relation is in between. For the Kuroshio and Oyashio Extension region, the most eddy-rich region in the North Pacific, eddies are responsible for over 50% of the total SSH variability up to the intra-seasonal scale, and ITD and HCT variability up to interannual. Eddy-induced SST variability is the highest along the Oyashio Extension Front on the order of 40–60% on submonthly scales. These results highlight the role of mesoscale eddies in ocean thermodynamic variability and in air-sea interaction.

Polarimetric and Electrical Structure of the 19 May 2013 Edmond–Carney, Oklahoma, Tornadic Supercell

AbstractWe demonstrate the utility of transient polarimetric signatures (ZDR and KDP columns, a proxy for surges in a thunderstorm updraft) to explain variability in lightning flash rates in a tornadic supercell. Observational data from a WSR-88D and the Oklahoma lightning mapping array are used to map the temporal variance of polarimetric signatures and VHF sources from lightning channels. It is shown, via three-dimensional and cross-sectional analyses, that the storm was of inverted polarity resulting from anomalous electrification. Statistical analysis confirms that mean flash area in the ZDR column region was 10 times smaller than elsewhere in the storm. On an average, 5 times more flash initiations occurred within ZDR column regions, thereby supporting existing theory of an inverse relationship between flash initiation rates and lightning channel extent. Segmentation and object identification algorithms are applied to gridded radar data to calculate metrics such as height, width, and volume of ZDR and KDP columns. Variability in lightning flash rates is best explained by the fluctuations in ZDR column volume with a Spearman’s rank correlation coefficient value of 0.72. The highest flash rates occur in conjunction with the deepest ZDR columns (up to 5 km above environmental melting level) and largest volumes of ZDR columns extending up to the −20°C level (3 km above the melting level). Reduced flash rates toward the end of the analysis are indicative of weaker updrafts manifested as low ZDR column volumes at and above the −10°C level. These findings are consistent with recent studies linking lightning to the interplay between storm dynamics, kinematics, thermodynamics, and precipitation microphysics.

Automatic analysis of social media images to identify disaster type and infer appropriate emergency response

Social media postings are increasingly being used in modern days disaster management. Along with the textual information, the contexts and cues inherent in the images posted on social media play an important role in identifying appropriate emergency responses to a particular disaster. In this paper, we proposed a disaster taxonomy of emergency response and used the same taxonomy with an emergency response pipeline together with deep-learning-based image classification and object identification algorithms to automate the emergency response decision-making process. We used the card sorting method to validate the completeness and correctness of the disaster taxonomy. We also used VGG-16 and You Only Look Once (YOLO) algorithms to analyze disaster-related images and identify disaster types and relevant cues (such as objects that appeared in those images). Furthermore, using decision tables and applied analytic hierarchy processes (AHP), we aligned the intermediate outputs to map a disaster-related image into the disaster taxonomy and determine an appropriate type of emergency response for a given disaster. The proposed approach has been validated using Earthquake, Hurricane, and Typhoon as use cases. The results show that 96% of images were categorized correctly on disaster taxonomy using YOLOv4. The accuracy can be further improved using an incremental training approach. Due to the use of cloud-based deep learning algorithms in image analysis, our approach can potentially be useful to real-time crisis management. The algorithms along with the proposed emergency response pipeline can be further enhanced with other spatiotemporal features extracted from multimedia information posted on social media.

Deep learning techniques‐based perfection of multi‐sensor fusion oriented human‐robot interaction system for identification of dense organisms

For detection of dense small-target organisms with indistinct features in complex background, the efficiency and accuracy of traditional target detection methods are low. Multi-sensor fusion oriented human-robot interaction (HRI) system has facilitated biologists to process and analyse data. For this, several deep learning models based on convolutional neural network (CNN) are improved and compared to study the species and density of dense organisms in deep-sea hydrothermal vent, which are fused it with related environmental information given by position sensors and conductivity-temperature-depth (CTD) sensors, so as to perfect multi-sensor fusion oriented HRI system. Firstly, the authors combined different meta-architectures and different feature extractors, and obtained five object identification algorithms based on CNN. Then, they compared computational cost of feature extractors and weighed the pros and cons of each algorithm from mean detection speed, correlation coefficient and mean class-specific confidence score to confirm that Faster Region-based CNN (R-CNN)_InceptionNet is the best algorithm applicable to hydrothermal vent biological dataset. Finally, they calculated the cognitive accuracy of rimicaris exoculata in dense and sparse areas, which were 88.3% and 95.9% respectively, to analyse the performance of the Faster R-CNN_InceptionNet. Results show that the proposed method can be used in the multi-sensor fusion oriented HRI system for the statistics of dense organisms in complex environments.

The construction of local convex hull on the task of pattern recognition

In this research paper analyzes and explores the problems of pattern recognition methods and biological object identification algorithms. We encountered two different problems in the process. The first is the modernization of the pattern recognition algorithm; the second is associated with the identification of a biological object. The algorithm problem was solved by constructing local convex hull during modernization. In the second, the Orthoptera insect group was selected as a biological object on the recommendation of experts. Orthoptera is a taxonomic insect division that includes locusts and other winged insects. Approaches to determining the characteristics of orthoptera types and preparing their samples, as well as the formation of identification models, are recommended. The algorithm for calculating estimates (ACE) was chosen when identifying the choice of orthoptera. The ACE based on the principle of partial precedence. Approaches to the formation of a training-test selection for the choice of orthopters are proposed. As mentioned above, in this work is proposed to develop an information recognition system designed to solve the problems associated with the identification of Orthoptera Insecta, as well as for the experimental study of the specie of the locust family.

Study on image processing method to classify objects on dynamic conveyor

Controlling a robotic arms for applications such as detection and classification moving object using the vision sensor is a trend in the field of industrial robots. In particular, the vision sensor is the "eye" of the robot. To solve this problem, we need an efficient image processing algorithm for object identification to optimize the speed. Our classification principle based on the color of the object to be classified first, then separating contour to classify according to the shape of the object. In addition, our paper also propose a classification method that rarely mentioned in the relevant documents that classify based on object's characteristic. In fact, the product packaging not only has one color, but also includes complex color and patterns. Being able to classify these products shows the practicality of the proposed method. For complex colors and patterns object, the PCASIFT algorithm is useful, where SIFT extracts the local characteristics of the object and PCA reduces the number of dimensionality and retain only the best characteristics for identification. To picking object, a proposed design with the optimal requirements of picking order so that picking time is the shortest to minimize the delay for the next picking. The other outstanding advantage is a system of robotic arm to perform pick-up and sorting. This helps to verify good running algorithms in real time. The items are randomly released and the rotation of items is random. The speed of the conveyor is 5cm/s, an average of more than 2 seconds to pick up an object and robot arm processing precisely at high speed. The experimental results using camera Logitech C270, Yamaha Scara YK-400X robotic arm, LabVolt conveyor and OpenCV library are satisfactory, reliable and applicable.