Car Dataset Research Articles

Multi-layer feature fusion and attention enhancement for fine-grained vehicle recognition research

Abstract Vehicle recognition technology is widely applied in automatic parking, traffic restrictions, and public security investigations, playing a significant role in the construction of intelligent transportation systems. Fine-grained vehicle recognition seeks to surpass conventional vehicle recognition by concentrating on more detailed sub-classifications. This task is more challenging due to the subtle inter-class differences and significant intra-class variations. Localization-classification subnetworks represent an efficacious approach frequently employed for this task, but previous research has typically relied on CNN deep feature maps for object localization, which suffer from the low resolution, leading to poor localization accuracy. The multi-layer feature fusion localization method proposed by us fuses the high-resolution feature map of the shallow layer of CNN with the deep feature map, and makes full use of the rich spatial information of the shallow feature map to achieve more precise object localization. In addition, traditional methods acquire local attention information through the design of complex models, frequently resulting in regional redundancy or information omission. To address this, we introduce an attention module that adaptively enhances the expressiveness of global features and generates global attention features. These global attention features are then integrated with object-level features and local attention cues to achieve a more comprehensive attention enhancement. Lastly, we devise a multi-branch model and employ the aforementioned object localization and attention enhancement methods for end-to-end training to make the multiple branches collaborate seamlessly to adequately extract fine-grained features. Extensive experiments conducted on the Stanford Cars dataset and the self-built Cars-126 dataset have demonstrated the effectiveness of our method, achieving a leading position among existing methods with 97.7% classification accuracy on the Stanford Cars dataset.

Protect your data and I’ll rank its utility: A framework for utility analysis of anonymized mobility data for smart city applications

When designing smart cities’ building blocks, mobility data plays a fundamental role in applications and services. However, mobility data usually comes with unrestricted location of its corresponding entities (e.g., citizens and vehicles) and poses privacy concerns, among them recovering the identity of those entities with linking attacks. Location Privacy Protection Mechanisms (LPPMs) based on anonymization, such as mix-zones, have been proposed to address the privacy of users’ identity. Once the data is protected, a comprehensive discussion about the trade-off between privacy and utility happens. However, issues still arise about the application of anonymized data to smart city development: what are the smart cities applications and services that can best leverage mobility data anonymized by mix-zones? To answer this question, we propose the Utility Analysis Framework of Anonymized Trajectories for Smart Cities-Application Domains (UAFAT). This characterization framework measures the utility through twelve metrics related to privacy, mobility, and social, including mix-zones performance metrics from anonymized trajectories produced by mix-zones. This framework aims to identify applications and services where the anonymized data will provide more or less utility in various aspects. The results evaluated with cabs and privacy cars datasets showed that further characterizing it by distortion level, UAFAT ranked the smart cities application domains that best leverage mobility data anonymized by mix-zones. Also, it identified which one of the four case studies of smart city applications had more utility. Additionally, different datasets present different behaviors in terms of utility. These insights can contribute significantly to the utility of both open and private data markets for smart cities.

Unsupervised Anomaly Detection for Power Batteries: A Temporal Convolution Autoencoder Framework

Abstract To prevent potential abnormalities from escalating into critical faults, a rapid and precise algorithm should be employed for detecting power battery anomalies. An unsupervised model based on a temporal convolutional autoencoder was proposed. It can quickly and accurately identify abnormal power battery data. Its encoder utilized a temporal convolutional network (TCN) structure with residuals to parallelly process data while capturing time dependencies. A novel TCN structure with an effect–cause relationship was developed for the decoder. The same-timescale connection was established between the encoder and decoder to improve the model performance. The validity of the proposed model was confirmed using a real-world car dataset. Compared with the GRU-AE model, the proposed approach reduced the parameter count and mean square error by 19.5% and 71.9%, respectively. This study provides insights into the intelligent battery pack abnormality detection technology.

Physically‐Based Lighting for 3D Generative Models of Cars

AbstractRecent work has demonstrated that Generative Adversarial Networks (GANs) can be trained to generate 3D content from 2D image collections, by synthesizing features for neural radiance field rendering. However, most such solutions generate radiance, with lighting entangled with materials. This results in unrealistic appearance, since lighting cannot be changed and view‐dependent effects such as reflections do not move correctly with the viewpoint. In addition, many methods have difficulty for full, 360° rotations, since they are often designed for mainly front‐facing scenes such as faces. We introduce a new 3D GAN framework that addresses these shortcomings, allowing multi‐view coherent 360° viewing and at the same time relighting for objects with shiny reflections, which we exemplify using a car dataset. The success of our solution stems from three main contributions. First, we estimate initial camera poses for a dataset of car images, and then learn to refine the distribution of camera parameters while training the GAN. Second, we propose an efficient Image‐Based Lighting model, that we use in a 3D GAN to generate disentangled reflectance, as opposed to the radiance synthesized in most previous work. The material is used for physically‐based rendering with a dataset of environment maps. Third, we improve the 3D GAN architecture compared to previous work and design a careful training strategy that allows effective disentanglement. Our model is the first that generate a variety of 3D cars that are multi‐view consistent and that can be relit interactively with any environment map.

GLC-Frame: A Framework and Library for Exploration of Multidimensional Data with General Line Coordinates

General Line Coordinates (GLC) are a relatively new set of line-based representations for visualizing multidimensional data with the distinctive characteristics of being reversible and lossless. Given these characteristics, the GLC have a high potential for exploratory multidimensional data analysis, however only partial implementations of some of the GLC techniques are available for the visualization community. In this paper, we present the GLC-Frame, an online exploration tool that supports a dual view and allows users to upload their own dataset and interactively explore the different GLC representations without writing code. We also present the GLC-Vis Library, an open-source data visualization library supporting GLC along with traditional interactions. Finally, we provide a set of usage examples showing how the different techniques behave in both the occlusion and the cluster identification problem. In addition, we present the interactions on GLC representations using the cars dataset. Both the GLC-Frame and the GLC-Vis Library provide an exploration space that will allow the visualization community to use these new techniques and evaluate their potential.

Sentiment Analysis of Self Driving Car Dataset: A comparative study of Deep Learning approaches

Sentiment Analysis (SA) is a crucial task in understanding public opinions and perceptions towards emerging technologies. In this study, we focus on SA for a self-driving car dataset as it provides valuable insights into public perceptions and opinions towards a transformative technology. The dataset consists of textual reviews associated with sentiment labels, providing insights into how people perceive self-driving car technology. Our objective is to analyze the sentiments expressed in these reviews using Deep Learning (DL) models, namely, Artificial Neural Network (ANN), Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), Gated Recurrent Unit (GRU), and Bidirectional GRU (BiGRU). We compared our results with an existing technique in the field of self-driving car sentiment classification, that implemented various Machine Learning (ML) and DL models, including Support Vector Machines (SVM), Naïve Bayes (NB), Logistic Regression (LR), Random Forest (RF), CNN, and LSTM. In our study, we expanded upon this research by evaluating the performance of ANN, BiLSTM, GRU, and BiGRU. Results reveal that BiLSTM, GRU, and BiGRU exhibit superior performance in sentiment classification within the self-driving car dataset. These findings offer valuable insights into public sentiment towards self-driving cars, contributing significantly to the advancement of SA techniques in the domain of autonomous vehicles. Additionally, the results are statistically tested and are statistically significant.

TransIFC: Invariant Cues-aware Feature Concentration Learning for Efficient Fine-grained Bird Image Classification

Fine-grained bird image classification (FBIC) is not only meaningful for endangered bird observation and protection but also a prevalent task for image classification in multimedia processing and computer vision. However, FBIC suffers from several challenges, such as bird molting, complex background, and arbitrary bird posture. To effectively tackle these challenges, we present a novel invariant cues-aware feature concentration Transformer (TransIFC), which learns invariant and core information in bird images. To this end, two novel modules are proposed to leverage the characteristics of bird images, namely, the hierarchy stage feature aggregation (HSFA) module and the feature in feature abstraction (FFA) module. The HSFA module aggregates the multiscale information of bird images by concatenating multilayer features. The FFA module extracts the invariant cues of birds through feature selection based on discrimination scores. Transformer is employed as the backbone to reveal the long-dependent semantic relationships in bird images. Moreover, abundant visualizations are provided to prove the interpretability of the HSFA and FFA modules in TransIFC. Comprehensive experiments demonstrate that TransIFC can achieve state-of-the-art performance on the CUB-200-2011 dataset (91.0%) and the NABirds dataset (90.9%). Finally, extended experiments have been conducted on the Stanford Cars dataset to suggest the potential of generalizing our method on other fine-grained visual classification tasks.

A hybrid model for energy-efficient Green Internet of Things enabled intelligent transportation systems using federated learning

In the rapidly evolving Internet of Things domain, managing voluminous data streams while ensuring energy efficiency remains a cardinal challenge. Our research introduces a hybrid model designed specifically for an IoT system. This paper sheds light on our unique framework that emphasizes efficient communication and prioritizes energy conservation. The innovative model architecture is lightweight, fostering compatibility with resource-constrained devices and paving the way for personalized federated averaging during training. A crucial highlight of our methodology includes local training using lightweight optimization instead of traditional data transmission, reducing energy overheads considerably. Furthermore, we propose a novel approach for model aggregation using personalized energy-aware averaging. This process iteratively refines a global model by aggregating received updates, drastically reducing the data transfer compared to conventional methods. Lastly, we integrate an energy-aware updates management system, continually monitoring device energy metrics and making adaptive data transmission and model update decisions. The model ensures optimal participation in global updates by consistently monitoring devices’ energy metrics and setting adaptive thresholds. This study leverages the richly annotated Udacity Self-Driving Car Dataset provided by Roboflow to evaluate a novel federated learning model to optimize energy consumption for IoT systems. Our experiments simulate real-world collaborative learning scenarios using the TensorFlow Federated (TFF). Our focus on energy consumption evaluation revealed that our proposed model offers significant energy savings when analyzed based on data communication round time and individual node training duration. A comparative analysis between the proposed and traditional models uncovers substantial improvements in energy efficiency. In our experiments, the proposed approach demonstrated a commendable accuracy of 93.27%. Notably, the local communication time was streamlined to 1.21 s, while the global communication was efficiently clocked at 4.76 s. When compared to traditional methods, these results not only underscore the effectiveness and efficiency of our methodology in the context of IoT systems but also highlight its superior performance.

Discrepant Semantic Diffusion Boosts Transfer Learning Robustness

Transfer learning could improve the robustness and generalization of the model, reducing potential privacy and security risks. It operates by fine-tuning a pre-trained model on downstream datasets. This process not only enhances the model’s capacity to acquire generalizable features but also ensures an effective alignment between upstream and downstream knowledge domains. Transfer learning can effectively speed up the model convergence when adapting to novel tasks, thereby leading to the efficient conservation of both data and computational resources. However, existing methods often neglect the discrepant downstream–upstream connections. Instead, they rigidly preserve the upstream information without an adequate regularization of the downstream semantic discrepancy. Consequently, this results in weak generalization, issues with collapsed classification, and an overall inferior performance. The main reason lies in the collapsed downstream–upstream connection due to the mismatched semantic granularity. Therefore, we propose a discrepant semantic diffusion method for transfer learning, which could adjust the mismatched semantic granularity and alleviate the collapsed classification problem to improve the transfer learning performance. Specifically, the proposed framework consists of a Prior-Guided Diffusion for pre-training and a discrepant diffusion for fine-tuning. Firstly, the Prior-Guided Diffusion aims to empower the pre-trained model with the semantic-diffusion ability. This is achieved through a semantic prior, which consequently provides a more robust pre-trained model for downstream classification. Secondly, the discrepant diffusion focuses on encouraging semantic diffusion. Its design intends to avoid the unwanted semantic centralization, which often causes the collapsed classification. Furthermore, it is constrained by the semantic discrepancy, serving to elevate the downstream discrimination capabilities. Extensive experiments on eight prevalent downstream classification datasets confirm that our method can outperform a number of state-of-the-art approaches, especially for fine-grained datasets or datasets dissimilar to upstream data (e.g., 3.75% improvement for Cars dataset and 1.79% improvement for SUN dataset under the few-shot setting with 15% data). Furthermore, the experiments of data sparsity caused by privacy protection successfully validate our proposed method’s effectiveness in the field of artificial intelligence security.

DEEP LEARNING FOR OBJECT DETECTION USING RADAR DATA

Abstract. Recently, Deep learning algorithms are becoming increasingly instrumental in autonomous driving by identifying and acknowledging road entities to ensure secure navigation and decision-making. Autonomous car datasets play a vital role in developing and evaluating perception systems. Nevertheless, the majority of current datasets are acquired using Light Detection and Ranging (LiDAR) and camera sensors. Utilizing deep neural networks yields remarkable outcomes in object recognition, especially when applied to analyze data from cameras and LiDAR sensors which perform poorly under adverse weather conditions such as rain, fog, and snow due to the sensor wavelengths. This paper aims to evaluate the ability to use RADAR dataset for detecting objects in adverse weather conditions, when LiDAR and Cameras may fail to be effective. This paper presents two experiments for object detection using Faster-RCNN architecture with Resnet-50 backbone and COCO evaluation metrics. Experiment 1 is object detection over only one class, while Experiment 2 is object detection over eight classes. The results show that as expected the average precision (AP) of detecting one class is (47.2) which is better than the results from detecting eight classes (27.4). Comparing my results from experiment 1 to the literature results which achieved an overall AP (45.77), my result was slightly better in accuracy than the literature mainly due to hyper-parameters optimization. The outcomes of object detection and recognition based on RADAR indicate the potential effectiveness of RADAR data in automotive applications particularly in adverse weather conditions, where vision and LiDAR may encounter limitations.

A Dynamic Path Optimization Model of IOT Delivery Vehicles for E-commerce Logistics Distribution

Logistics and distribution is a vital link to guarantee the stable supply of the e-commerce market and the healthy development of the industry. With the constant growth of the e-commerce, the efficiency and service quality of logistics and distribution have been paid more and more attention to. Therefore, the study firstly Considering distribution fixed cost, transportation penalty cost and carbon emission cost, the vehicle routing optimization model is transformed into the lowest transportation cost model, then uses an improved traditional artificial fish swarm algorithm to find the optimum way for this model, and finally verifies its performance and applicability through experiments. The performance test results show that the algorithm finds the optimal solution 3589 and 3590 in 63 and 78 iterations in the Oxford Robot Car dataset and Apollo Scape dataset, respectively; the average running time of the algorithm is 11.864s and 11.967s in the 10 operation time tests; in the operation function test, the algorithm. The algorithm was able to overcome the local optimal solution problem. The applicability simulation shows that this algorithm stabilizes after 53 iterations, the minimum cost of the optimal solution of the model is $41,224, and the total distance of distribution is 9035 km. The research algorithm is fast in finding the optimal value, which is close to it, indicating that the algorithm is highly efficient and reliable, and can greatly optimize the path of e-commerce logistics delivery vehicles, and give a theoretical foundation for the optimization of logistics delivery paths in other industries.

An Improved Big Data Analytics Architecture Using Federated Learning for IoT-Enabled Urban Intelligent Transportation Systems

The exponential growth of the Internet of Things has precipitated a revolution in Intelligent Transportation Systems, notably in urban environments. An ITS leverages advancements in communication technologies and data analytics to enhance the efficiency and intelligence of transport networks. At the same time, these IoT-enabled ITSs generate a vast array of complex data classified as Big Data. Traditional data analytics frameworks need help to efficiently process these Big Data due to its sheer volume, velocity, variety, and significant data privacy concerns. Federated Learning, known for its privacy-preserving attributes, is a promising technology for implementation within ITSs for IoT-generated Big Data. Nevertheless, the system faces challenges due to the variable nature of devices, the heterogeneity of data, and the dynamic conditions in which ITS operates. Recent efforts to mitigate these challenges focus on the practical selection of an averaging mechanism during the server’s aggregation phase and practical dynamic client training. Despite these efforts, existing research still relies on personalized FL with personalized averaging and client training. This paper presents a personalized architecture, including an optimized Federated Averaging strategy that leverages FL for efficient and real-time Big Data analytics in IoT-enabled ITSs. Various personalization methods are applied to enhance the traditional averaging algorithm. Local fine-tuning and weighted averaging tailor the global model to individual client data. Custom learning rates are utilized to boost the performance further. Regular evaluations are advised to maintain model efficacy. The proposed architecture addresses critical challenges like real-life federated environment settings, data integration, and significant data privacy, offering a comprehensive solution for modern urban transportation systems using Big Data. Using the Udacity Self-Driving Car Dataset foe vehicle detection, we apply the proposed approaches to demonstrate the efficacy of our model. Our empirical findings validate the superiority of our architecture in terms of scalability, real-time decision-making capabilities, and data privacy preservation. We attained accuracy levels of 93.27%, 92.89%, and 92.96% for our proposed model in a Federated Learning architecture with 10 nodes, 20 nodes, and 30 nodes, respectively.

Supervised contrastive ResNet and transfer learning for the in-vehicle intrusion detection system

High-end vehicles have been equipped with several electronic control units (ECUs), which provide upgrading functions to enhance the driving experience. The controller area network (CAN) is a well-known protocol that connects these ECUs because of its modesty and efficiency. However, the CAN bus is vulnerable to various types of attacks. Although the intrusion detection system (IDS) is proposed to address the security problem of the CAN bus, most previous studies only provide alerts when attacks occur without knowing the specific type of attack. Moreover, an IDS is designed for a specific car model due to diverse car manufacturers. In this study, we proposed a novel deep learning model called supervised contrastive (SupCon) ResNet, which can handle multiple attack classification on the CAN bus. Furthermore, the model can be used to improve the performance of a limited-size dataset using a transfer learning technique. The capability of the proposed model is evaluated on two real car datasets. When tested with the Car Hacking dataset, the experiment results show that the SupCon loss reduces the overall false-negative rates of four types of attack by an average of five times compared to the vanilla cross-entropy loss. In addition, the model achieves the highest F1 score on both the vehicle models of the survival dataset by utilizing transfer learning. Finally, the model can adapt to hardware constraints in terms of memory size and running time to be deployed in real devices.

An Optimal BDCNN ML Architecture for Car Make Model Prediction

We take on the challenge of classifying car photos, from the most general car type to the precise make, model, and year of the vehicle for a given input. Analyzing pre-existing datasets, we find that the CompCars-SV are a great place to begin our classification project. We demonstrate that convolutional neural networks can obtain a classification accuracy of more than 90% on the most difficult task. Due to a skewed mix between training and testing, this impressive result isn't really typical of how people do in the actual world. Using an ML system for car detection, we automatically generate a vehicle-tight bounding box for each picture, which we disseminate to the full dataset together with the existing (but limited) type-level annotation. We have designed and implemented car classification algorithms to analyze this car dataset, two of which take advantage of the hierarchical nature of car annotations. According to our research, a more precise classification of car type at a finer resolution now achieves an accuracy of 99.25%. It serves as a baseline benchmark for future research. Focusing on "vehicle" tasks, this work intends to bring attention to the vision community's lack of attention to these tasks compared to other objects. The important reason getting higher accuracy is extraction of binary descriptor (BD) feature using edge detection before training the CNN. This step reduced the size of the car dataset; hence network took less time to get trained. From the result outcomes shown it is clear that the presented network architecture having 31 layers of 2d convolutional layer, batch normalization, maxpool, ReLU, fully connected layer and Softmax classifier layer, has given higher accuracy. Numerous relevant car-related issues and solutions have yet to be carefully examined and researched, according to our findings. Car model categorization, model verification, and attribute prognosis are just a few examples of how the dataset might be put to use.

Fine-grained Image Recognition via Attention Interaction and Counterfactual Attention Network

Learning subtle and discriminative regions plays an important role in fine-grained image recognition, and attention mechanisms have shown great potential in such tasks. Recent research mainly focuses on employing the attention mechanism to locate key discriminative regions and learn salient features, whilst ignoring imperceptible complementary features and the causal relationship between prediction results and attention. To address the above issues, we propose an Attention Interaction and Counterfactual Attention Network (AICA-Net). Specifically, we propose an Attention Interaction Fusion Module (AIFM) to model the negative correlation between the attention map channels to locate the complementary features, and fuse the complementary features and key discriminative features to generate richer fine-grained features. Simultaneously, an Enhanced Counterfactual Attention Module (ECAM) is proposed to generate a counterfactual attention map. By comparing the impact of the learned attention map and the counterfactual attention map on the final prediction results, quantifying the quality of attention drives the network to learn more effective attention. Extensive experiments on CUB-200-2011, FGVC-Aircraft and Stanford Cars datasets have shown that our AICA-Net can get outstanding results. In particular, it achieves 90.83% and 95.87% accuracy on two open competitive benchmark datasets CUB-200-2011 and Stanford Cars, respectively. Experiments demonstrate that our method outperforms state-of-the-art solutions.

Fine-Grained Image Recognition by Means of Integrating Transformer Encoder Blocks in a Robust Single-Stage Object Detector

Fine-grained image classification remains an ongoing challenge in the computer vision field, which is particularly intended to identify objects within sub-categories. It is a difficult task since there is both minimal and substantial intra-class variance. Current methods address the issue through first locating selective regions with region proposal networks (RPNs), object localization, or part localization, followed by implementing a CNN network or SVM classifier to those selective regions. This approach, however, makes the process simple via implementing a single-stage end-to-end feature encoded with a localization method, which leads to improved feature representations of individual tokens/regions through integrating the transformer encoder blocks into the Yolov5 backbone structure. These transformer encoder blocks, with their self-attention mechanism, effectively capture global dependencies and enable the model to learn relationships between distant regions. This improves the model’s ability to understand context and capture long-range spatial relationships in an image. We also replaced the Yolov5 detection heads with three transformer heads at the output for object recognition using the discriminative and informative feature maps from transformer encoder blocks. We established the potential of the single-stage detector for the fine-grained image recognition task, achieving state-of-the-art 93.4% accuracy, as well as outperforming existing one-stage recognition models. The effectiveness of our approach is assessed using the Stanford car dataset, which includes 16,185 images of 196 different classes of vehicles with significantly identical visual appearances.

Multi-branch selection fusion fine-grained classification algorithm based on coordinate attention localization

Object localization has been the focus of research in Fine-Grained Visual Categorization (FGVC). With the aim of improving the accuracy and precision of object localization in multi-branch networks, as well as the robustness and universality of object localization methods, our study mainly focus on how to combine coordinate attention and feature activation map for target localization. The model in this paper is a three-branch model including raw branch, object branch and part branch. The images are fed directly into the raw branch. Coordinate Attention Object Localization Module (CAOLM) is used to localize and crop objects in the image to generate the input for the object branch. Attention Partial Proposal Module (APPM) is used to propose part regions at different scales. The three classes of input images undergo end-to-end weakly supervised learning through different branches of the network. The model expands the receptive field to capture multi-scale features by Selective Branch Atrous Spatial Pooling Pyramid (SB-ASPP). It can fuse the feature maps obtained from the raw branch and the object branch with Selective Branch Block (SBBlock), and the complete features of the raw branch are used to supplement the missing information of the object branch. Extensive experimental results on CUB-200-2011, FGVC-Aircraft and Stanford Cars datasets show that our method has the best classification performance on FGVC-Aircraft and also has competitive performance on other datasets. Few parameters and fast inference speed are also the advantages of our model.

Used Car Price Prediction by Using XGBoost

This article demonstrates that by using methods such as Extreme Gradient Boosting (XGBoost), dummy variables, etc., the selling price can be accurately predicted according to the different conditions and variables of each used car. The used car dataset is divided into a training dataset and a test dataset according to the ratio of 83% and 17%. This article uses a total of three data processing methods to find the most accurate prediction method. The first is to remove the outliers of the training dataset and test dataset, and then directly use the xgboost prediction method for prediction. The second is to remove the outliers and remove the variable power that is most closely related to the price of the used car, and then use the xgboost prediction method to make predictions. The third method is to remove outliers and then normalize the training dataset and test dataset, finally using the xgboost prediction method to predict. The experimental results show that normalizing the dataset and then using XGBoost and dummy variables can be used to predict the selling price accurately and efficiently through the different usage conditions of each used car.

A Self-Supervised Tree-Structured Framework for Fine-Grained Classification

In computer vision, fine-grained classification has become an important issue in recognizing objects with slight visual differences. Usually, it is challenging to generate good performance when solving fine-grained classification problems using traditional convolutional neural networks. To improve the accuracy and training time of convolutional neural networks in solving fine-grained classification problems, this paper proposes a tree-structured framework by eliminating the effect of differences between clusters. The contributions of the proposed method include the following three aspects: (1) a self-supervised method that automatically creates a classification tree, eliminating the need for manual labeling; (2) a machine-learning matcher which determines the cluster to which an item belongs, minimizing the impact of inter-cluster variations on classification; and (3) a pruning criterion which filters the tree-structured classifier, retaining only the models with superior classification performance. The experimental evaluation of the proposed tree-structured framework demonstrates its effectiveness in reducing training time and improving the accuracy of fine-grained classification across various datasets in comparison with conventional convolutional neural network models. Specifically, for the CUB 200 2011, FGVC aircraft, and Stanford car datasets, the proposed method achieves a reduction in training time of 32.91%, 35.87%, and 14.48%, and improves the accuracy of fine-grained classification by 1.17%, 2.01%, and 0.59%, respectively.

Car Emotion Labeling Based on Color-SSL Semi-Supervised Learning Algorithm by Color Augmentation

In the era of emotional consumption, it has become a hot topic that commodities meet consumers’ emotional needs. As a necessity of life, the car also needs to meet the needs of consumers. To achieve that consumers can purchase cars according to their emotional needs, we need to label cars with emotional words. The car’s appearance is the crucial medium of emotional information transmission, especially the car’s color is an essential emotional factor. As the first impression of products, color affects people’s emotional attitude. Therefore, introducing color features into the training process of sample marking is an excellent idea for intelligent labeling of a large number of product emotions. This paper proposes a semi-supervised learning method, Color-SSL, based on color data augmentation to realize the label of car emotion. Color-SSL takes FlexMatch as the framework of a semi-supervised learning model and augments data by extracting subject color. Compared with the baseline method, the accuracy of this method improved by 3.2%, 8.3%, 8.6%, and 1.4% with 10, 50, 100, and 200 training samples and 1000 test samples. The results show that Color-SSL obtains the best emotion-label result (94%). In addition, this study publishes pictures of emotional car datasets with high resolution, orthogonal perspective, and uniform background.