Object Features Research Articles

Feature Enhancement Based Oriented Object Detection in Remote Sensing Images

Since objects in remote sensing imagery often have arbitrary orientations and high densities, the features of small objects are inclined to be contaminated by the background and other instances. To address the issues, we propose a new oriented object detection framework where a series of feature enhancement schemes are implemented so as to improve robustness and accuracy of the detector. Firstly, we design a weighted bidirectional feature pyramid network, which can be used to fuse both high-level semantic features and low-level detail features for effectively handling with multi-scale objects. Accordingly, we apply the convolutional block attention module that exploits both spatial- and channel-wise attention in our detector, and study how to effectively integrate it into the framework for adaptive feature refinement. In the meanwhile, we present a semantic segmentation guided module to generate naive mask, which is used to multiple with pyramid features to filter out background noise and improve feature representation for small objects. The experimental results on two public datasets, i.e., UCAS-AOD and DOTA, validate the effective performance of the proposed method for oriented object detection in remote sensing images.

The Mathematical Model Based on the Parameters of Simulation Results Predicts the Fire Extinguishing Resource Demand of Naval Fires

In recent years, due to the diversity of fire scenes in ports and ships, the problem of fire command is complicated. In the case of power failure, the original command platform and fire extinguishing system will not be able to be used normally, or the fire extinguishing resources carried by the ship when it is on fire cannot be used. In the traditional firefighting ability research, there is no clear measure of firefighting ability, resulting in inaccurate calculation results. By corresponding to the quantity and composition of resources for the study of fire extinguishing capability, the combination of the fire dynamics simulation software PyroSim2020 and a calculation model that converts its resources into the total amount and flow of fire extinguishing agent can be provided. Based on the analytic hierarchy process (AHP), a firefighting demand grade calculation and judgment model was established, which included five factors: fire scale, combustion material characteristics, foam extinguishing agent performance, firefighting object characteristics, and external factors. It concluded that the demand assignment values of each index element were 2, 5, 6, 9, and 10, and further proposed the grade judgment criteria for the calculation results of comprehensive fire demand. Through the quantitative simulation of firefighting demand based on the fire scenario and calculation model test of the consumption prediction of cooperative firefighting equipment, it can also provide a strategic reference for related cooperative fire rescue.

Adaptation of Object Detection Algorithms for Road Indicator Lights in Complex Scenes

In the realm of autonomous driving, practical driving scenarios are fraught with numerous complexities, including inclement weather conditions, nighttime blurriness, and ambient light sources that significantly hinder drivers’ ability to discern road indicators. Furthermore, the dynamic nature of road indicators, which are constantly evolving, poses additional challenges for computer vision-based detection systems. To address these issues, this paper introduces a road indicator light detection model, leveraging the advanced capabilities of YOLOv8. We have ingeniously integrated the robust backbone of YOLOv8 with four distinct attention mechanism modules—Convolutional Block Attention Module (CBAM), Efficient Channel Attention (ECA), Shuffle Attention (SA), and Global Attention Mechanism (GAM)—to significantly enhance the model performance in capturing nuanced features of road indicators and boosting the accuracy of detecting minute objects. Additionally, we employ the Asymptotic Feature Pyramid Network (AFPN) strategy, which optimizes the fusion of features across different scales, ensuring not only an enhanced performance but also maintaining real-time capability. These innovative attention modules empower the model by recalibrating the significance of both channel and spatial dimensions within the feature maps, enabling it to hone in on the most pertinent object characteristics. To tackle the challenges posed by samples rich in small, occluded, background-similar objects, and those that are inherently difficult to recognize, we have incorporated the Focaler-IOU loss function. This loss function deftly reduces the contribution of easily detectable samples to the overall loss, thereby intensifying the focus on challenging samples. This strategic balancing of hard-to-detect versus easy-to-detect samples effectively elevates the model’s detection performance. Experimental evaluations conducted on both a public traffic signal dataset and a proprietary headlight dataset have yielded impressive results, with both mAP50 and mAP50:95 metrics experiencing significant improvements exceeding two percentage points. Notably, the enhancements observed in the headlight dataset are particularly profound, signifying a significant step forward toward realizing safer and more reliable assisted driving technologies.

Ordinal information, but not metric information, matters in binding feature with depth location in three-dimensional contexts.

A basic function of human visual perception is the ability to recognize and locate objects in the environment. It has been shown that two-dimensional (2D) location can reliably bias judgments on object identity (spatial congruency bias; Golomb et al., 2014), suggesting that 2D location information is automatically bound with object features to induce such a bias. Although the binding problem of feature and location has been vigorously studied under various 2D settings, it remains unclear how depth location can be bound with object features in a three-dimensional (3D) setting. Here we conducted five experiments in various 3D contexts using the congruency bias paradigm, and found that changes of object's depth location could influence perceptual judgments on object features differently depending on whether its relative depth order with respect to others changed or not. Experiments 1 and 2 showed that the judgments on an object's color could be affected by changes in its ordinal depth, but not by changes in its absolute metric depth. Experiment 3 showed that the bias was asymmetric-changes in an object's color did bias the judgments on metric-depth location, but not if its depth order had changed. Experiments 4 and 5 investigated whether these findings could be generalized to a peripersonal near space and a large-scale far space, respectively, using more ecological virtual environments. Our findings suggest that ordinal depth plays a special role in feature-location binding: an object may be automatically bound with its relative depth relation, but not with its absolute metric-depth location. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

HOIMotion: Forecasting Human Motion During Human-Object Interactions Using Egocentric 3D Object Bounding Boxes.

We present HOIMotion - a novel approach for human motion forecasting during human-object interactions that integrates information about past body poses and egocentric 3D object bounding boxes. Human motion forecasting is important in many augmented reality applications but most existing methods have only used past body poses to predict future motion. HOIMotion first uses an encoder-residual graph convolutional network (GCN) and multi-layer perceptrons to extract features from body poses and egocentric 3D object bounding boxes, respectively. Our method then fuses pose and object features into a novel pose-object graph and uses a residual-decoder GCN to forecast future body motion. We extensively evaluate our method on the Aria digital twin (ADT) and MoGaze datasets and show that HOIMotion consistently outperforms state-of-the-art methods by a large margin of up to 8.7% on ADT and 7.2% on MoGaze in terms of mean per joint position error. Complementing these evaluations, we report a human study (N=20) that shows that the improvements achieved by our method result in forecasted poses being perceived as both more precise and more realistic than those of existing methods. Taken together, these results reveal the significant information content available in egocentric 3D object bounding boxes for human motion forecasting and the effectiveness of our method in exploiting this information.

Efficient Indoor 3D Object Detection in Point Clouds Using the Kinect Sensor

Abstract Only using the point clouds collected by a Kinect sensor to perform 3D object detection is a very challenging task for indoor sence understanding. In this paper, we propose a novel point clouds based 3D object detection framework named Weighted VoteNet (W-VoteNet), which systematically enhances seed points, voting points, and proposal objects to varying degree, thereby enhancing the detection accuracy of indoor scene point clouds. Firstly, a new neighborhood feature enhancement module was designed to enhance the features of the seed point, which can better utilize the neighborhood feature of the seed point. Secondly, a weighted voting module is implemented to increase the accuracy of voting, allowing more prospect seed points to participate in voting. Finally, a new semantic relation reasoning module is proposed to get the semantic relationship features of the proposal object, which can further decrease false alarm rate. The three modules we proposed all contribute to more accurate voting and more effective proposals. The experimental results on two benchmark indoor 3D object detection datasets, SUN RGB-D and ScanNet V2, demonstrate the effectiveness of our approach. The source code is available at: https://github.com/Guo-JQ/W-VoteNet

Research on Land Use and Land Cover Information Extraction Methods for Remote Sensing Images Based on Improved Convolutional Neural Networks

To address the challenges that convolutional neural networks (CNNs) face in extracting small objects and handling class imbalance in remote sensing imagery, this paper proposes a novel spatial contextual information and multiscale feature fusion encoding–decoding network, SCIMF-Net. Firstly, SCIMF-Net employs an improved ResNeXt-101 deep backbone network, significantly enhancing the extraction capability of small object features. Next, a novel PMFF module is designed to effectively promote the fusion of features at different scales, deepening the model’s understanding of global and local spatial contextual information. Finally, introducing a weighted joint loss function improves the SCIMF-Net model’s performance in extracting LULC information under class imbalance conditions. Experimental results show that compared to other CNNs such as Res-FCN, U-Net, SE-U-Net, and U-Net++, SCIMF-Net improves PA by 0.68%, 0.54%, 1.61%, and 3.39%, respectively; MPA by 2.96%, 4.51%, 2.37%, and 3.45%, respectively; and MIOU by 3.27%, 4.89%, 4.2%, and 5.68%, respectively. Detailed comparisons of locally visualized LULC information extraction results indicate that SCIMF-Net can accurately extract information from imbalanced classes and small objects.

AFA–Mamba: Adaptive Feature Alignment with Global–Local Mamba for Hyperspectral and LiDAR Data Classification

The joint classification of hyperspectral imagery (HSI) and LiDAR data is an important task in the field of remote sensing image interpretation. Traditional classification methods, such as support vector machine (SVM) and random forest (RF), have difficulty capturing the complex spectral–spatial–elevation correlation information. Recently, important progress has been made in HSI-LiDAR classification using Convolutional Neural Networks (CNNs) and Transformers. However, due to the large spatial extent of remote sensing images, the vanilla Transformer and CNNs struggle to effectively capture global context. Moreover, the weak misalignment between multi-source data poses challenges for their effective fusion. In this paper, we introduce AFA–Mamba, an Adaptive Feature Alignment Network with a Global–Local Mamba design that achieves accurate land cover classification. It contains two main core designs: (1) We first propose a Global–Local Mamba encoder, which effectively models context through a 2D selective scanning mechanism while introducing local bias to enhance the spatial features of local objects. (2) We also propose an SSE Adaptive Alignment and Fusion (A2F) module to adaptively adjust the relative positions between multi-source features. This module establishes a guided subspace to accurately estimate feature-level offsets, enabling optimal fusion. As a result, our AFA–Mamba consistently outperforms state-of-the-art multi-source fusion classification approaches across multiple datasets.

Research on Deep Learning Detection Model for Pedestrian Objects in Complex Scenes Based on Improved YOLOv7

Objective: Pedestrian detection is very important for the environment perception and safety action of intelligent robots and autonomous driving, and is the key to ensuring the safe action of intelligent robots and auto assisted driving. Methods: In response to the characteristics of pedestrian objects occupying a small image area, diverse poses, complex scenes and severe occlusion, this paper proposes an improved pedestrian object detection method based on the YOLOv7 model, which adopts the Convolutional Block Attention Module (CBAM) attention mechanism and Deformable ConvNets v2 (DCNv2) in the two Efficient Layer Aggregation Network (ELAN) modules of the backbone feature extraction network. In addition, the detection head is replaced with a Dynamic Head (DyHead) detector head with an attention mechanism; unnecessary background information around the pedestrian object is also effectively excluded, making the model learn more concentrated feature representations. Results: Compared with the original model, the log-average miss rate of the improved YOLOv7 model is significantly reduced in both the Citypersons dataset and the INRIA dataset. Conclusions: The improved YOLOv7 model proposed in this paper achieved good performance improvement in different pedestrian detection problems. The research in this paper has important reference significance for pedestrian detection in complex scenes such as small, occluded and overlapping objects.

Explaining wine scores through stochastic frontier analysis. An intent to measure sensory non-observable attributes

ABSTRACT Experts give scores to wines, which are quality proxies for marketers and buyers. The production of scores proxying wine quality is explained by a set of observable objective attributes, plus another set of unobservable and subjective (sensory) features, and randomness. Is it possible to use a Stochastic Frontier Approach (SFA) to understand whether objective and subjective (sensory) characteristics of wines explain the differences in ‘producing’ wine scores? We estimate a wine quality stochastic frontier production function, using a database of 1800 top-scored wines, in 18 years-window encompassing objective determinants that largely explain the scores (price, production, year, grape, country, etcetera), being aware that sensory aspects related to wine grading are unobservable. We find that the observable variables explain part of the ‘efficiency’ in attaining scores and suggest that ‘inefficiency’ in score generation can be attributed to unobservable (sensory) variables.

Flowering Index Intelligent Detection of Spray Rose Cut Flowers Using an Improved YOLOv5s Model

Addressing the current reliance on manual sorting and grading of spray rose cut flowers, this paper proposed an improved YOLOv5s model for intelligent recognition and grading detection of rose color series and flowering index of spray rose cut flowers. By incorporating small-scale anchor boxes and small object feature output, the model enhanced the annotation accuracy and the detection precision for occluded rose flowers. Additionally, a convolutional block attention module attention mechanism was integrated into the original network structure to improve the model’s feature extraction capability. The WIoU loss function was employed in place of the original CIoU loss function to increase the precision of the model’s post-detection processing. Test results indicated that for two types of spray rose cut flowers, Orange Bubbles and Yellow Bubbles, the improved YOLOv5s model achieved an accuracy and recall improvement of 10.2% and 20.0%, respectively. For randomly collected images of spray rose bouquets, the model maintained a detection accuracy of 95% at a confidence threshold of 0.8.

Survey Defining Lip Health-Dermatology and Health Care Providers' Perspectives.

There currently exists no consensus on objective features which are relevant in the assessment of overall lip health. This study seeks to identify features and factors that are associated with healthy lips. This survey will ultimately enable clinicians to objectively track lip health for clinical evaluations and clinical research. An anonymous survey was conducted among allied healthcare professionals in-person and through email. The survey contained questions designed to assess the importance of features of lip health using a 5-point Likert scale. Features included lack of lip dryness; even tone or lack of hyperpigmentation; lip smoothness; lip color; lip shine, radiance, or luster; definition of the vermillion border/defined lip contour; lip fullness or plumpness; lack of lip lines; and lip firmness. The final questions prompted respondents to select the top three most important features when assessing lip health and to provide any additional pertinent factors. A total of 334 responses were received, including board-certified dermatologists (35.9%), dermatology residents (14.7%), and other healthcare specialties (49.4%). The features most cited were lack of lip dryness (65.8%), lip smoothness (38.4%), even tone or lack of hyperpigmentation (38.1%), definition of the vermillion border/defined lip contour (28.0%). Survey results reveal that lip dryness, lip smoothness, lack of hyperpigmentation, and definition of the vermillion border/defined lip contour were the features most associated with lip health. The results from this survey will provide a basis for future research in the field of lip health, research, and clinical treatment.

YOLO-DHGC: Small Object Detection Using Two-Stream Structure with Dense Connections

Small object detection, which is frequently applied in defect detection, medical imaging, and security surveillance, often suffers from low accuracy due to limited feature information and blurred details. This paper proposes a small object detection method named YOLO-DHGC, which employs a two-stream structure with dense connections. Firstly, a novel backbone network, DenseHRNet, is introduced. It innovatively combines a dense connection mechanism with high-resolution feature map branches, effectively enhancing feature reuse and cross-layer fusion, thereby obtaining high-level semantic information from the image. Secondly, a two-stream structure based on an edge-gated branch is designed. It uses higher-level information from the regular detection stream to eliminate irrelevant interference remaining in the early processing stages of the edge-gated stream, allowing it to focus on processing information related to shape boundaries and accurately capture the morphological features of small objects. To assess the effectiveness of the proposed YOLO-DHGC method, we conducted experiments on several public datasets and a self-constructed dataset. Exceptionally, a defect detection accuracy of 96.3% was achieved on the Market-PCB public dataset, demonstrating the effectiveness of our method in detecting small object defects for industrial applications.

Joint Optimization-Based Texture and Geometry Enhancement Method for Single-Image-Based 3D Content Creation

Recent studies have explored the generation of three-dimensional (3D) meshes from single images. A key challenge in this area is the difficulty of improving both the generalization and detail simultaneously in 3D mesh generation. To address this issue, existing methods utilize fixed-resolution mesh features to train networks for generalization. This approach is capable of generating the overall 3D shape without limitations on object categories. However, the generated shape often exhibits a blurred surface and suffers from suboptimal texture resolution due to the fixed-resolution mesh features. In this study, we propose a joint optimization method that enhances geometry and texture by integrating generalized 3D mesh generation with adjustable mesh resolution. Specifically, we apply an inverse-rendering-based remeshing technique that enables the estimation of complex-shaped mesh estimations without relying on fixed-resolution structures. After remeshing, we enhance the texture to improve the detailed quality of the remeshed mesh via a texture enhancement diffusion model. By separating the tasks of generalization, detailed geometry estimation, and texture enhancement and adapting different target features for each specific network, the proposed joint optimization method effectively addresses the characteristics of individual objects, resulting in increased surface detail and the generation of high-quality textures. Experimental results on the Google Scanned Objects and ShapeNet datasets demonstrate that the proposed method significantly improves the accuracy of 3D geometry and texture estimation, as evaluated by the PSNR, SSIM, LPIPS, and CD metrics.

SMART ATTENDANCE WITH FACE ANTI-SPOOFING TECHNOLOGY USING HAAR CASCADE CLASSIFIER

Traditional attendance systems often encounter challenges in efficiently and accurately recording attendance. This research aims to introduce an innovative solution through the development of an intelligent anti-spoofing attendance system based on facial recognition using the Haar Cascade Classifier method. Designed to overcome the inefficiencies in attendance recording, this system ensures the accuracy of educational staff attendance records. Its development method relies on the Haar Cascade Classifier, employing image processing to detect learned object features, particularly focusing on facial recognition. Research findings indicate that the implementation of this system achieves an average accuracy rate of 98.90% in attendance recording. The facial recognition technology ensures reliable attendance recording with confidence levels exceeding 80%, signifying precise facial identification that addresses various challenges and ensures attendance data integrity. Not only does the system identify educational staff with high accuracy, but it also provides prompt responses for efficient attendance logging and verification. Beyond its technical benefits, this study significantly contributes to the development of smarter and more efficient attendance technology. The system plays a crucial role in enhancing the discipline of educational staff at STMIK IKMI Cirebon and streamlining attendance management and evaluation across educational institutions.

Charting the growth through intelligence: A SWOC analysis on AI-assisted radiologic bone age estimation.

Bone age estimation (BAE) is based on skeletal maturity and degenerative process of the skeleton. The clinical importance of BAE is in understanding the pediatric and growth-related disorders; whereas medicolegally it is important in determining criminal responsibility and establishing identification. Artificial Intelligence (AI) has been used in the field of the field of medicine and specifically in diagnostics using medical images. AI can greatly benefit the BAE techniques by decreasing the intra observer and inter observer variability as well as by reducing the analytical time. The AI techniques rely on object identification, feature extraction and segregation. Bone age assessment is the classical example where the concepts of AI such as object recognition and segregation can be used effectively. The paper describes various AI based algorithms developed for the purpose of radiologic BAE and the performances of the models. In the current paper we have also carried out qualitative analysis using Strength, Weakness, Opportunities and Challenges (SWOC) to examine critical factors that contribute to the application of AI in BAE. To best of our knowledge, the SWOC analysis is being carried out for the first time to assess the applicability of AI in BAE. Based on the SWOC analysis we have provided strategies for successful implementation of AI in BAE in forensic and medicolegal context.

Fast feature- and category-related parafoveal previewing support free visual exploration.

While humans typically saccade every ∼250 ms in natural settings, studies on vision tend to prevent or restrict eye movements. As it takes ∼50 ms to initiate and execute a saccade, this leaves only ∼200 ms to identify the fixated object and select the next saccade goal. How much detail can be derived about parafoveal objects in this short time interval, during which foveal processing and saccade planning both occur? Here, we had male and female human participants freely explore a set of natural images while we recorded magnetoencephalography and eye movements. Using multivariate pattern analysis, we demonstrate that future parafoveal images could be decoded at the feature and category level with peak decoding at ∼110 ms and ∼165 ms respectively, while the decoding of fixated objects at the feature and category level peaked at ∼100 ms and ∼145 ms. The decoding of features and categories was contingent on the objects being saccade goals. In sum, we provide insight on the neuronal mechanism of pre-saccadic attention by demonstrating that feature and category specific information of foveal and parafoveal objects can be extracted in succession within a ∼200 ms intersaccadic interval. These findings rule out strict serial or parallel processing accounts but are consistent with a pipeline mechanism in which foveal and parafoveal objects are processed in parallel but at different levels in the visual hierarchy.Significance Statement We provide neural evidence that future parafoveal saccade goals are processed surprisingly quickly at the feature and the category level before we saccade to them. Specifically, using multivariate pattern analysis applied to magnetoencephalography and eye-tracking data, we found that information about the colour and the category of parafoveal objects emerged at ∼110 ms and ∼165 ms respectively, with the same information about foveal objects emerging ∼100 ms and ∼145 ms. Our findings provide novel insight into the neuronal dynamics of parafoveal previewing during free visual exploration. The dynamics rule out strict serial or parallel processing, but are consistent with a pipelining mechanism in which foveal and parafoveal objects are processed in parallel but at different levels in the visual hierarchy.

Enhancing quadrotor robustness control using image-based visual servoing (IBVS) with fuzzy logic

Visual servoing is a commonly employed approach in robotics and unmanned aerial vehicles (UAVs) that facilitates accurate object positioning and movement control by utilizing visual feedback. As quadrotors gain popularity, more control methods have been developed. This article presents a method for enhancing quadrotor robustness control using image-based visual servoing (IBVS) with fuzzy logic. Unlike traditional visual servoing, which relies on a fixed gain and often encounters challenges with velocity convergence and maintaining the object in the field of view, this method is designed to enhance the visual servo control of quadrotors by dynamically adjusting the gain of the IBVS system through a fuzzy logic controller. This controller adaptively adjusts the servo gain in response to feature errors and the depth of the object's feature points. MATLAB simulations clearly demonstrate the superior performance of this fuzzy logic integrated method compared to classical approaches, showcasing enhanced control capabilities in challenging environments.

Ecological Capacity and Bioenergy Potential of Agrocenoses as a Way of Sustainable Agriculture

Background: Calculation of bioenergy potential and ecological capacity are promising methods for assessing the state of an agro-landscape, allowing us to promptly respond to changes in its sustainability. The article presents the results of soil research and calculation of bioenergy potential of the farm territory in the zone of dry steppe on chestnut soils. Methods: Assessment of anthropogenic load on agricultural landscape is based on ecological standards and point system of relative fertility. The determination of the ecological capacity by BEPT of an agrocenosis occupied by winter wheat and sunflower was carried out according to the method of V.M. Volodin, R.F. Eremina, N.F. Mikhailova. Determination of erosion and geomorphological characteristics of the research object was carried out using geoinformation technologies. The basis for creation of digital elevation model (DEM) is the data of radar survey of the Earth surface (SRTM4). Result: Soils of the research object have low and average humus content. Nitrogen level in the soil varies from very low to good; nitrogen availability in the fields is average. Provision with mobile phosphorus is average. Soil potassium content is mostly high. As a result of calculations, it was found that in the conditions of the dry steppe zone, the bioenergy potential of the agrolandscape area occupied by winter wheat is 2735.72 GJ/ha, sunflower-2647.4 GJ/ha. At the same time, the ecological capacity of the field occupied by winter wheat is 2764.6 GJ/ha and sunflower-2676.8 GJ/h.

YOLO-ESL: An Enhanced Pedestrian Recognition Network Based on YOLO

Pedestrian detection is a critical task in computer vision; however, mainstream algorithms often struggle to achieve high detection accuracy in complex scenarios, particularly due to target occlusion and the presence of small objects. This paper introduces a novel pedestrian detection algorithm, YOLO-ESL, based on the YOLOv7 framework. YOLO-ESL integrates the ELAN-SA module, designed to enhance feature extraction, with the LGA module, which improves feature fusion. The ELAN-SA module optimizes the flexibility and efficiency of small object feature extraction, while the LGA module effectively integrates multi-scale features through local and global attention mechanisms. Additionally, the CIOUNMS algorithm addresses the issue of target loss in cases of high overlap, improving boundary box filtering. Evaluated on the VOC2012 pedestrian dataset, YOLO-ESL achieved an accuracy of 93.7%, surpassing the baseline model by 3.0%. Compared to existing methods, this model not only demonstrates strong performance in handling occluded and small object detection but also remarkable robustness and efficiency.