V3 Network Research Articles

A Machine Learning and Deep Learning-Based Account Code Classification Model for Sustainable Accounting Practices

Accounting account codes are created within a specific logic framework to systematically and accurately record a company’s financial transactions. Currently, accounting reports are processed manually, which increases the likelihood of errors and slows down the process. This study aims to use image processing techniques to predict cash codes in accounting reports, automate accounting processes, improve accuracy, and save time. Deep learning embeddings from Inception V3, SqueezeNet, VGG-19, VGG-16, Painters, and DeepLoc networks were utilized in the feature extraction phase. A total of six learning algorithms, namely Logistic Regression, Gradient Boosting, Neural Network, kNN, Naive Bayes, and Stochastic Gradient Descent were employed to classify the images. The highest accuracy rate of 99.2% was achieved with the combination of the Inception V3 feature extractor and the Neural Network classifier. The results demonstrate that image processing methods significantly reduce error rates in accounting records, accelerate processes, and support sustainable accounting practices. This indicates that image processing techniques have substantial potential to contribute to digital transformation in accounting, helping businesses achieve their sustainability goals.

Research on Radar Target Detection Based on the Electromagnetic Scattering Imaging Algorithm and the YOLO Network

In this paper, the radar imaging technology based on the time-domain (TD) electromagnetic scattering algorithm is used to generate image datasets quickly and apply them to target detection research. Considering that radar images are different from optical images, this paper proposes an improved strategy for the traditional You Only Look Once (YOLO)v3 network to improve target detection accuracy on radar images. The speckle noise in radar images can cover the real information of a target image and increase the difficulty of target detection. The attention mechanisms are added to the traditional YOLOv3 network to strengthen the weight of the target region. By comparing the target detection accuracy under different attention mechanisms, an attention module with higher detection accuracy is obtained. The validity of the proposed detection network is verified on a simulation dataset, a measured real dataset, and a mixed dataset. This paper is about an interdisciplinary study of computational electromagnetics, remote sensing, and artificial intelligence. Experiments verify that the proposed composite network has better detection performance.

Research on the Method of Imperfect Wheat Grain Recognition Utilizing Hyperspectral Imaging Technology

As the primary grain crop in China, wheat holds a significant position in the country’s agricultural production, circulation, consumption, and various other aspects. However, the presence of imperfect grains has greatly impacted wheat quality and, subsequently, food security. In order to detect perfect wheat grains and six types of imperfect grains, a method for the fast and non-destructive identification of imperfect wheat grains using hyperspectral images was proposed. The main contents and results are as follows: (1) We collected wheat grain hyperspectral data. Seven types of wheat grain samples, each containing 300 grains, were prepared to construct a hyperspectral imaging system for imperfect wheat grains, and visible near-infrared hyperspectral data from 2100 wheat grains were collected. The Savitzky–Golay algorithm was used to analyze the hyperspectral images of wheat grains, selecting 261 dimensional effective hyperspectral datapoints within the range of 420.61–980.43 nm. (2) The Successive Projections Algorithm was used to reduce the dimensions of the 261 dimensional hyperspectral datapoints, selecting 33 dimensional hyperspectral datapoints. Principal Component Analysis was used to extract the optimal spectral wavelengths, specifically selecting hyperspectral images at 647.57 nm, 591.78 nm, and 568.36 nm to establish the dataset. (3) Particle Swarm Optimization was used to optimize the Support Vector Machines model, Convolutional Neural Network model, and MobileNet V2 model, which were established to recognize seven types of wheat grains. The comprehensive recognition rates were 93.71%, 95.14%, and 97.71%, respectively. The results indicate that a larger model with more parameters may not necessarily yield better performance. The research shows that the MobileNet V2 network model exhibits superior recognition efficiency, and the integration of hyperspectral image technology with the classification model can accurately identify imperfect wheat grains.

Improved ShuffleNet V2 network with attention for speech emotion recognition

This paper enhanced human-robot interaction (HRI) towards interacting with humans through a speech-emotion recognition algorithm based on the improved ShuffleNet V2 network as the means of communication. When recognizing speech emotion, previous studies typically analyze speech signals solely from the time or frequency domain, resulting in the loss of detailed information. Our algorithm initially converts speech data into an acoustic spectrogram to address this limitation, preserving rich original information across both time and frequency domains. This approach optimizes feature extraction by leveraging the capabilities of deep convolutional networks to capture local changes in images. Furthermore, we integrated the attention mechanism from ECA-Net into the ShuffleNet V2 model. This incorporation enhances the extraction of significant features, amplifying the expression of features strongly correlated with the emotional state conveyed in speech while suppressing irrelevant ones. In addition to these enhancements, we replace the ReLU activation function with Hardswish to deepen the neural network, thereby improving feature information without compromising model accuracy. Finally, the training stage incorporates the hot restart Cosine Annealing Learning Rate to optimize the network further. Simulation experiments are conducted on the enterface'05 dataset to evaluate the accuracy of the proposed algorithm. The experimental results show that the recognition rate of the method is higher than that of the existing algorithm in terms of the effectiveness of HRI.

Enhanced Image Captioning for Social Media using Inception V3 and Transformer Networks

Enhanced Image Captioning for Social Media using Inception V3 and Transformer Networks

Augmenting interpretation of vaginoscopy observations in cycling bitches with deep learning model

Successful identification of estrum or other stages in a cycling bitch often requires a combination of methods, including assessment of its behavior, exfoliative vaginal cytology, vaginoscopy, and hormonal assays. Vaginoscopy is a handy and inexpensive tool for the assessment of the breeding period. The present study introduces an innovative method for identifying the stages in the estrous cycle of female canines. With a dataset of 210 vaginoscopic images covering four reproductive stages, this approach extracts deep features using the inception v3 and Residual Networks (ResNet) 152 models. Binary gray wolf optimization (BGWO) is applied for feature optimization, and classification is performed with the extreme gradient boosting (XGBoost) algorithm. Both models are compared with the support vector machine (SVM) with the Gaussian and linear kernel, k-nearest neighbor (KNN), and convolutional neural network (CNN), based on performance metrics such as accuracy, specificity, F1 score, sensitivity, precision, matthew correlation coefficient (MCC), and runtime. The outcomes demonstrate the superiority of the deep model of ResNet 152 with XGBoost classifier, achieving an average model accuracy of 90.37%. The method gave a specific accuracy of 90.91%, 96.38%, 88.37%, and 88.24% in predicting the proestrus, estrus, diestrus, and anestrus stages, respectively. When performing deep feature analysis using inception v3 with the same classifiers, the model achieved an accuracy of 89.41%, which is comparable to the results obtained with the ResNet model. The proposed model offers a reliable system for identifying the optimal mating period, providing breeders and veterinarians with an efficient tool to enhance the success of their breeding programs.

IScan: Detection of Colorectal Cancer from CT Scan Images Using Deep Learning

Colorectal cancer, a highly lethal form of cancer, can be treated effectively if detected early. However, the current diagnosis process involves a time-consuming and manual review of CT scans to identify cancerous regions and behavior, leading to resource consumption, subjectivity, and dependency on manual assessment. We propose a 3-phase deep neural system for automated colorectal cancer detection using CT scan images to address these challenges. It includes a SegNet network to identify tumor locations, an InceptionResNet V2 network to classify tumors as benign or malignant, and an analysis of tumor area cum perimeter to predict the cancer stage. The proposed model offers a fully automated solution by combining these functionalities under a single umbrella. In real-life CT scans from 37 patients, the proposed model achieved 95.8% ROI segmentation accuracy, a dice coefficient of 0.6214, 69.75% IoU score, and 95.83% tumor classification accuracy. The unique approach using Radial Length (RL) and Circularity (C) parameters predicted the T-stage with close to 85% accuracy. Based on these outcomes, the proposed system establishes itself as a reliable and suitable alternative to traditional cancer diagnosis techniques by leveraging the power of automation, deep learning, and innovative parameter analysis.

Enhancing mobile robot navigation: integrating reactive autonomy through deep learning and fuzzy behavior

Objective: This study aimed to develop a control architecture for reactive autonomous navigation of a mobile robot by integrating Deep Learning techniques and fuzzy behaviors based on traffic signal recognition. Materials: The research utilized transfer learning with the Inception V3 network as a base for training a neural network to identify traffic signals. The experiments were conducted using a Donkey-Car, an Ackermann-steering-type open-source mobile robot, with inherent computational limitations. Results: The implementation of the transfer learning technique yielded a satisfactory result, achieving a high accuracy of 96.2% in identifying traffic signals. However, challenges were encountered due to delays in frames per second (FPS) during testing tracks, attributed to the Raspberry Pi's limited computational capacity. Conclusions: By combining Deep Learning and fuzzy behaviors, the study demonstrated the effectiveness of the control architecture in enhancing the robot's autonomous navigation capabilities. The integration of pre-trained models and fuzzy logic provided adaptability and responsiveness to dynamic traffic scenarios. Future research could focus on optimizing system parameters and exploring applications in more complex environments to further advance autonomous robotics and artificial intelligence technologies.

Application of Multi-objective Evolutionary Algorithms for Multidimensional Sensory Data Prediction and Resource Scheduling in Smart City Design

Multidimensional sensory data prediction and resource scheduling are paramount challenges in the design of smart cities. This paper delves into the utilization of multi-objective evolutionary algorithms to enhance the accuracy and efficiency of target detection through optimized YOLO_v3 network models. By integrating the YOLO_v3 model with the K-means++ algorithm for Anchor_Box generation, the novel approach exhibits superior adaptability and flexibility, particularly in handling variable-sized feature pattern mappings. This adaptability better caters to the detection of targets of diverse sizes, thus elevating the performance and precision of target detection algorithms. To further scrutinize the YOLO-v3 joint algorithm’s performance in urban traffic detection, P-R curves were plotted for various loss types on the NEU-DET dataset. Comparative analysis of these curves highlights the optimized algorithms’ superiority in detecting various types of losses in urban model completeness. Additionally, practical application analysis revealed that the optimized monitoring results outperform the detection time of the original YOLO-v3_means++ network model on FP_GA. Notably, post-processing with C-FENCE can reduce average single-frame image detection time to 2.01 seconds, while convolutional degree-level fusion with the BN layer cuts it down to 2.25 seconds. In summary, the FP_GA-based YOLO-v3_means++ network algorithm offers superior detection capabilities, and the multi-objective evolutionary algorithm’s optimization of the YOLO-v3 model enhances target detection performance and precision.

IAE-KM3D a 3D Object Detection Method Based on an Improved KM3D Network

Deep learning-based 3D target detection methods need to solve the problem of insufficient 3D target detection accuracy. In this paper, the KM3D network is selected as the benchmark network after the experimental comparison of current mainstream algorithms, and the IAE-KM3D network algorithm based on the KM3D network is proposed. First, the Resnet V2 network is introduced, and the residual module is redesigned to improve the training capability of the new residual module with higher generalization. IBN NET is then introduced to carefully integrate instance normalization and batch normalization as building blocks to improve the model’s detection accuracy in hue- and brightness-changing scenarios without increasing time loss. Then, a parameter-free attention mechanism, Simam, is introduced to improve the detection accuracy of the model. After that, the elliptical Gaussian kernel is introduced to improve the algorithm’s ability to detect 3D targets. Finally, a new key point loss function is proposed to improve the algorithm’s ability to train. Experiments using the KITTI dataset conclude that the IAE-KM3D network model significantly improves detection accuracy and outperforms the KM3D algorithm regarding detection performance compared to the original KM3D network. The improvements for AP2D, AP3D, and APBEV are 5%, 12.5%, and 8.3%, respectively, and only a tiny amount of time loss and network parameters are added. Compared with other mainstream target detection algorithms, Monn3D, 3DOP, GS3D, and FQNet, the improved IAE-KM3D network in this paper significantly improves AP3D and APBEV, with fewer network parameters and shorter time consumption.

Citrus Pest Identification Model Based on Improved ShuffleNet

To address the current issues of complex structures and low accuracies in citrus pest identification models, a lightweight pest identification model was proposed. First, a parameterized linear rectification function was introduced to avoid neuronal death. Second, the model’s attention to pest characteristics was improved by incorporating an improved mixed attention mechanism. Subsequently, the network structure of the original model was adjusted to reduce architectural complexity. Finally, by employing transfer learning, an SCHNet model was developed. The experimental results indicated that the proposed model achieved an accuracy rate of 94.48% with a compact size of 3.84 MB. Compared to the original ShuffleNet V2 network, the SCHNet model showed a 3.12% accuracy improvement while reducing the model size by 22.7%. The SCHNet model exhibited an excellent classification performance for citrus pest identification, enabling the accurate identification of citrus pests.

Deep Learning-Based Lung Medical Image Recognition

Pulmonary nodules serve as critical indicators for early lung cancer diagnosis, making their detection and classification essential. The prevalent use of transfer learning in recognition algorithms often encounters a significant disparity between source and target datasets, which hampers effective feature extraction from pulmonary nodules and degrades performance. An enhanced neural network model leveraging convolutional neural networks is introduced to address this issue. This model integrates a pre-trained GoogLeNet Inception V3 network with a custom-designed feature fusion layer, improving the network’s ability to extract features. To ascertain the optimal configuration, the models were evaluated based on accuracy in various combinations. The experiments conducted on the LUNA16 pulmonary nodule dataset revealed that the refined network model achieved an accuracy of 88.78% and a sensitivity of 87.18%. This represents an increase of 2.7 and 2.22 percentage points in accuracy and sensitivity, respectively, compared to the GoogLeNet Inception V3 algorithm. Further tests across different dataset proportions also yielded superior outcomes, demonstrating enhanced generalization capabilities. These findings can offer objective benchmarks for clinical diagnosis.

An object detection system using array planar capacitive sensors

This paper presents a target detection system using array-based planar capacitive sensors, demonstrating excellent detection performance in gesture recognition. The system achieves gesture target classification by collecting the capacitance values of the planar capacitive sensor array and optimizing the BP neural network through a genetic algorithm for model training. Additionally, the system employs visual detection for gesture targets, using transfer learning with the MobileNet V2 network in the Keras framework to achieve gesture target classification. The system utilizes a data fusion mechanism to merge visual recognition and planar capacitive sensor array recognition for decision-making. Experimental results demonstrate that the target detection system exhibits outstanding detection performance in gesture recognition, with an average accuracy of up to 98.92%.

Evaluation of strength and stiffness degradation of RC shear walls: An integrated image processing and deep learning approach

AbstractIn the aftermath of an earthquake, damage detection and performance evaluation of structural components are imperative for assessing the residual seismic capacity of a building. In this study, an integrated image processing and deep learning approach was developed to evaluate the degradation in strength and stiffness (i.e., strength reduction and stiffness reduction) of reinforced concrete (RC) shear walls. The approach comprised two main tasks: detecting and localizing visible seismic damage from photographs and evaluating strength and stiffness degradation based on this information. The semantic segmentation network, Damage‐Net, was used for damage detection and localization. A novel crack morphological processing layer and a patch feature extraction layer were developed for damage feature extraction and compression. A lightweight deep convolutional neural network named DegradeEval‐Net_v2, featuring the upgraded dilated and separable convolution block and multi‐layer perception, was developed to link the damage feature with strength and stiffness degradation. A database comprising test data and photographs of 14 RC shear wall specimens with a flexural‐dominated behavior mode and high to intermediate ductility was constructed to train and test the DegradeEval‐Net_v2 network. The results indicate that DegradeEval‐Net_v2 substantially improved the performance assessment accuracy of damaged RC shear walls, with a 35% smaller root mean square error (RMSE) for stiffness degradation evaluation and 75% smaller RMSE for strength degradation evaluation, compared with the provisions specified in JBDPA and FEMA guidelines. Moreover, evaluation results on test sets demonstrate that introducing the damage feature extraction and compression layers effectively preserved local crack information and improved the accuracy with which stiffness reduction was evaluated. In addition, DegradeEval‐Net_v2 outperformed ResNet18 and MobileNet V3 in terms of balanced efficiency and accuracy. Interpretability analysis demonstrates that the model learned the distinct contribution patterns of various visible damage indexes to stiffness and strength degradation across different loading levels.

Enhanced YOLO v3 for precise detection of apparent damage on bridges amidst complex backgrounds

A bridge disease identification approach based on an enhanced YOLO v3 algorithm is suggested to increase the accuracy of apparent disease detection of concrete bridges under complex backgrounds. First, the YOLO v3 network structure is enhanced to better accommodate the dense distribution and large variation of disease scale characteristics, and the detection layer incorporates the squeeze and excitation (SE) networks attention mechanism module and spatial pyramid pooling module to strengthen the semantic feature extraction ability. Secondly, CIoU with better localization ability is selected as the loss function for training. Finally, the K-means algorithm is used for anchor frame clustering on the bridge surface disease defects dataset. 1363 datasets containing exposed reinforcement, spalling, and water erosion damage of bridges are produced, and network training is done after manual labelling and data improvement in order to test the efficacy of the algorithm described in this paper. According to the trial results, the YOLO v3 model has enhanced more than the original model in terms of precision rate, recall rate, Average Precision (AP), and other indicators. Its overall mean Average Precision (mAP) value has also grown by 5.5%. With the RTX2080Ti graphics card, the detection frame rate increases to 84 Frames Per Second, enabling more precise and real-time bridge illness detection.

Design and Experiment of Variable-Spray System Based on Deep Learning

In response to issues of the low pesticide-utilization rate caused by the traditional constant spraying method, a variable-spraying system for wheat-field weeds was proposed in this study based on real-time segmentation by deep learning. In this study, the weed density within the operational area was obtained by using the improved DeepLab V3+ semantic segmentation mode, and a variable spray-level model based on real-time weed density and speed was constructed by using PWM variable-spraying technology to adjust the spray volume. The lightweight MobileNet V2 network was selected as its backbone network, and the CA attention mechanism was integrated into the feature extraction module. The mean intersection over the union (MIoU) and mean pixel accuracy (MPA) of the improved DeepLab V3+ were 73.34% and 80.76%, respectively, and the segmentation time for a single image was 0.09 s. The results of field verification tests showed that (1) compared with constant spraying, variable spraying can increase droplet density and save the amount of pesticides, with the droplet density increased by 38.87 droplets/cm2 and the pesticide consumption saved by 46.3%; and (2) at the same speed, the average droplet-coverage rate in the areas with sparse weed density decreased by 13.98% compared with the areas with dense weeds. Under the same plant density, the average coverage rate of 0.5 m/s increased by 2.91% and 6.59% compared with 1 m/s and 1.5 m/s, respectively. These results further demonstrated that the system can automatically adjust the spray volume based on different travel speeds and weed densities. This research can provide theoretical and reference support for the development of new precision-spray plant-protection machinery for wheat fields.

An Open-Circuit Fault Diagnosis Method for Three-Level Neutral Point Clamped Inverters Based on Multi-Scale Shuffled Convolutional Neural Network.

This study constructs a power switching device open-circuit fault diagnosis model for a three-level neutral point clamped inverter based on the multi-scale shuffled convolutional neural network (MSSCNN) and extracts and classifies the fault information contained in the output current of inverters. The model employs depthwise separable convolution and channel shuffle techniques to improve diagnostic accuracy and reduce model complexity. The experimental results show that the new model has lower model complexity, better noise resistance and higher average diagnostic accuracy compared with fault diagnosis models based on CNN, ResNet, ShuffleNet V2 and Mobilenet V3 networks.

A severity estimation method for lightweight cucumber leaf disease based on DM-BiSeNet

Accurately estimating the severity of cucumber diseases is crucial for improving cucumber quality and minimizing economic losses. Deep learning techniques have shown promising results in automatically extracting disease image features for severity estimation. However, existing methods still face challenges in accurately estimating disease severity under complex backgrounds and achieving real-time performance.This paper presents a lightweight severity estimation method called DM-BiSeNet to address these challenges. The proposed method utilizes BiSeNet V2 as the base network and incorporates depthwise separable convolutional blocks to optimize the detail branch. A simplified MobileNet V3 network is also constructed to optimize the semantic branch. The model training process is accelerated using the AdamW optimizer. To evaluate the performance of DM-BiSeNet, a dataset consisting of cucumber powdery mildew and downy mildew disease images collected in natural scenes is utilized. Experimental results demonstrate that DM-BiSeNet achieves higher accuracy in severity estimation, with an R2 value of 0.9407 and an RMSE of 1.0680, outperforming the comparison methods. Moreover, DM-BiSeNet exhibits a complexity of 1.54 GFLOPs and is capable of reasoning 94 disease images per second.The proposed DM-BiSeNet model offers a lightweight and effective solution for accurate and rapid severity estimation of cucumber diseases under complex backgrounds. It provides a valuable technical tool for quantitative disease estimation, offering significant potential for practical applications.

Enhanced Knowledge Distillation for Advanced Recognition of Chinese Herbal Medicine.

The identification and classification of traditional Chinese herbal medicines demand significant time and expertise. We propose the dual-teacher supervised decay (DTSD) approach, an enhancement for Chinese herbal medicine recognition utilizing a refined knowledge distillation model. The DTSD method refines output soft labels, adapts attenuation parameters, and employs a dynamic combination loss in the teacher model. Implemented on the lightweight MobileNet_v3 network, the methodology is deployed successfully in a mobile application. Experimental results reveal that incorporating the exponential warmup learning rate reduction strategy during training optimizes the knowledge distillation model, achieving an average classification accuracy of 98.60% for 10 types of Chinese herbal medicine images. The model boasts an average detection time of 0.0172 s per image, with a compressed size of 10 MB. Comparative experiments demonstrate the superior performance of our refined model over DenseNet121, ResNet50_vd, Xception65, and EfficientNetB1. This refined model not only introduces an approach to Chinese herbal medicine image recognition but also provides a practical solution for lightweight models in mobile applications.

A CHERRY FRUIT RECOGNITION METHOD FOR CURRENT TASK BASED ON DEEP LEARNING

To improve the accuracy of cherry fruit recognition during picking operations and to enable automated cherry picking by robots, both original and complex sample image sets were collected at the orchard site in two phases. Concurrently, the basic data set, enhanced data set, and task data set were generated, wherein the recognition targets were all mature cherry fruits and the recognition targets were cherry fruits that became within the current operation range. First, the basic and enhanced data sets were used for comparative training on you only look once (YOLO) v3. The test results demonstrated that the ability of YOLO v3 to recognize occluded and overlapping cherry fruits can be greatly enhanced by increasing the quantity and percentage of samples in the training set that have high levels of occlusion and overlap. Second, the enhanced data set was used to train the Faster R-CNN, SSD, YOLO v3, and YOLO v5 networks. The test results indicated that YOLO v5 had the best recognition effect on mature cherries. Next, the enhanced data set and the task data set were used to train YOLO v5. Experiments revealed that YOLO v5 was more successful in identifying cherry fruits that needed to be picked within the current task range, while shielding mature cherries that should not be picked outside of it. 96% of the data was precise, and 99% of the data was recall.