Detection Network Research Articles

A lightweight network model designed for alligator gar detection

When using advanced detection algorithms to monitor alligator gar in real-time in wild waters, the efficiency of existing detection algorithms is subject to certain limitations due to turbid water quality, poor underwater lighting conditions, and obstruction by other objects. In order to solve this problem, we developed a lightweight real-time detection network model called ARD-Net, from the perspective of reducing the amount of calculation and obtaining more feature map patterns. We introduced a cross-domain grid matching strategy to accelerate network convergence, and combined the involution operator and dual-channel attention mechanism to build a more lightweight feature extractor and multi-scale detection reasoning network module to enhance the network’s response to different semantics. Compared with the yoloV5 baseline model, our method performs equivalently in terms of detection accuracy, but the model is smaller, the detection speed is increased by 1.48 times, When compared with the latest State-of-the-Art (SOTA) method, YOLOv8, our method demonstrates clear advantages in both detection efficiency and model size,and has good real-time performance. Additionally, we created a dataset of alligator gar images for training.

CSDNet: A Novel Deep Learning Framework for Improved Cataract State Detection

Cataracts, known for lens clouding and being a common cause of visual impairment, persist as a primary contributor to vision loss and blindness, presenting notable diagnostic and prognostic challenges. This work presents a novel framework called the Cataract States Detection Network (CSDNet), which utilizes deep learning methods to improve the detection of cataract states. The aim is to create a framework that is more lightweight and adaptable for use in environments or devices with limited memory or storage capacity. This involves reducing the number of trainable parameters while still allowing for effective learning of representations from data. Additionally, the framework is designed to be suitable for real-time or near-real-time applications where rapid inference is essential. This study utilizes cataract and normal images from the Ocular Disease Intelligent Recognition (ODIR) database. The suggested model employs smaller kernels, fewer training parameters, and layers to efficiently decrease the number of trainable parameters, thereby lowering computational costs and average running time compared to other pre-trained models such as VGG19, ResNet50, DenseNet201, MIRNet, Inception V3, Xception, and Efficient net B0. The experimental results illustrate that the proposed approach achieves a binary classification accuracy of 97.24% (normal or cataract) and an average cataract state detection accuracy of 98.17% (normal, grade 1—minimal cloudiness, grade 2—immature cataract, grade 3—mature cataract, and grade 4—hyper mature cataract), competing with state-of-the-art cataract detection methods. The resulting model is lightweight at 17 MB and has fewer trainable parameters (175, 617), making it suitable for deployment in environments or devices with constrained memory or storage capacity. With a runtime of 212 ms, it is well-suited for real-time or near-real-time applications requiring rapid inference.

A bearing surface defect detection method based on multi-attention mechanism Yolov8

Surface defects in bearings not only affect the appearance but also impact the service life and performance. Therefore, it is imperative for bearing manufacturers to conduct quality inspections before bearings leave the factory. However, traditional visual inspection methods exhibit shortcomings such as high omission rates, insufficient feature fusion and oversized models when dealing with multiple target defects in bearings. To address these challenges, this paper proposes a surface defect detection method for bearings based on an improved Yolov8 algorithm (G-Yolov8). Firstly, a C3Ghost convolutional module based on the Ghost module is constructed in YOLOv8 to simplify model computational costs. Secondly, a global attention mechanism module is designed at the end of the backbone network to increase sensitivity to implicit small target area features and optimize feature extraction efficiency. Subsequently, a deep deformable convolution feature pyramid network is constructed by introducing the deformable convolutional networks version 2 (DCNv2) and the lightweight content-aware reassembly of features upsampling operator to reduce sampling information loss and improve the fusion of multi-scale target defects. Finally, different attention mechanisms are embedded in the detection network to construct a multi-attention detection head to replace the decoupled head, refining classification and localization tasks, reducing feature confusion, and improving the model’s detection accuracy. Experimental results demonstrate that the improved algorithm achieves a 3.5% increase in mean average precision on a self-made small-scale train bearing surface defect dataset, with a 17.3% reduction in model size. This improvement not only enhances accuracy but also addresses the requirement for lightweight deployment in subsequent stages.

Receptive field enhancement and attention feature fusion network for underwater object detection

Receptive field enhancement and attention feature fusion network for underwater object detection

End-to-End lightweight Transformer-Based neural network for grasp detection towards fruit robotic handling

Robotic picking and placing are common operations for fruits and vegetables in grading, sorting or packaging systems. However, due to the diverse shapes and irregular surfaces of fruits and vegetables, improper handling during the picking process can result in detachment or damage. To ensure the correct grasping positions, it is necessary to design targeted neural network algorithms for achieving intelligent sorting. Therefore, this study focuses on 20 common fruit and vegetable agricultural products to develop a deep learning-based grasping detection algorithm model. By combining local features from convolutional neural networks with global features from Transformers, a lightweight end-to-end fruit and vegetable grasping detection network, MDETR, is constructed. Experimental results demonstrate that the MDETR algorithm not only achieves high accuracy in fruit and vegetable grasping detection but also improves the speed of pose detection. The average time required for detecting a single image is approximately 29.6 ms, meeting real-time requirements. The algorithm achieves a pose detection accuracy rate of 96 %, enabling precise detection and positioning of fruit and vegetable poses and achieving fast and accurate picking and placing. Additionally, a Pybullet simulation platform is developed for conducting grasping experiments, where the MDETR model achieves a grasping success rate of 88.9 %. This validates the robustness and generalization capabilities of the proposed detection algorithm model, designed specifically for fruit and vegetable grasping tasks.

Comparison of U-Net and Fully Convolutional Networks (FCN) for concrete cracks detection using raw images under various conditions

Crack detection in concrete buildings is crucial for assessing structural health, but it poses challenges due to complex backgrounds, real-time requirements, and high accuracy demands. Deep learning techniques, including U-Net and Fully Convolutional Networks (FCN), have shown promise in crack detection. However, they are sensitive to real-world environmental variations, impacting robustness and accuracy. This paper compares the performance of U-Net and FCN for concrete crack detection on bridges using raw images under various conditions. A dataset of 157 images (100 for training, 57 for testing) was used, and the models were evaluated based on Dice similarity coefficient and Jaccard index. FCN slightly outperformed U-Net in accuracy (94.88% vs. 94.21%), while U-Net had a slight advantage in validation (93.55% vs. 92.99%). These findings provide valuable insights for automated infrastructure maintenance and repair.

Artificial intelligence to automate assessment of ocular and periocular measurements.

To develop and validate a deep learning facial landmark detection network to automate the assessment of periocular anthropometric measurements. Patients presenting to the ophthalmology clinic were prospectively enrolled and had their images taken using a standardised protocol. Facial landmarks were segmented on the images to enable calculation of marginal reflex distance (MRD) 1 and 2, palpebral fissure height (PFH), inner intercanthal distance (IICD), outer intercanthal distance (OICD), interpupillary distance (IPD) and horizontal palpebral aperture (HPA). These manual segmentations were used to train a machine learning algorithm to automatically detect facial landmarks and calculate these measurements. The main outcomes were the mean absolute error and intraclass correlation coefficient. A total of 958 eyes from 479 participants were included. The testing set consisted of 290 eyes from 145 patients. The AI algorithm demonstrated close agreement with human measurements, with mean absolute errors ranging from 0.22 mm for IPD to 0.88 mm for IICD. The intraclass correlation coefficients indicated excellent reliability (ICC > 0.90) for MRD1, MRD2, PFH, OICD, IICD, and IPD, while HPA showed good reliability (ICC 0.84). The landmark detection model was highly accurate and achieved a mean error rate of 0.51% and failure rate at 0.1 of 0%. The automated facial landmark detection network provided accurate and reliable periocular measurements. This may help increase the objectivity of periocular measurements in the clinic and may facilitate remote assessment of patients with tele-health.

Novel Ransomware Detection Exploiting Uncertainty and Calibration Quality Measures Using Deep Learning

Ransomware poses a significant threat by encrypting files or systems demanding a ransom be paid. Early detection is essential to mitigate its impact. This paper presents an Uncertainty-Aware Dynamic Early Stopping (UA-DES) technique for optimizing Deep Belief Networks (DBNs) in ransomware detection. UA-DES leverages Bayesian methods, dropout techniques, and an active learning framework to dynamically adjust the number of epochs during the training of the detection model, preventing overfitting while enhancing model accuracy and reliability. Our solution takes a set of Application Programming Interfaces (APIs), representing ransomware behavior as input we call “UA-DES-DBN”. The method incorporates uncertainty and calibration quality measures, optimizing the training process for better more accurate ransomware detection. Experiments demonstrate the effectiveness of UA-DES-DBN compared to more conventional models. The proposed model improved accuracy from 94% to 98% across various input sizes, surpassing other models. UA-DES-DBN also decreased the false positive rate from 0.18 to 0.10, making it more useful in real-world cybersecurity applications.

Aka-Net: anchor free-based object detection network for surveillance video transmission in the IOT edge computing environment

Aka-Net: anchor free-based object detection network for surveillance video transmission in the IOT edge computing environment

Wavy-attention network for real-time cyber-attack detection in a small modular pressurized water reactor digital control system

Global interest in advanced reactors has been reignited by recent investments in small modular reactors and micro-reactor design. The use of digital devices is essential for meeting the size and modularity requirements of small modular reactor controls. By fully digitizing the small modular reactor control systems, critical information can be obtained to optimize control, reduce costs, and extend the reactor’s lifetime. However, the potential for cyber-attacks on digital devices leaves digital control systems vulnerable. To address this risk, this study presents a novel wavy-attention network for sensor attack detection in nuclear plants. The wavy-attention network comprises stacks of batch-normalized, dilated, one-dimensional convolution neural networks, and sequential self-attention modules, superior to conventional single-layer networks on sequence classification tasks. To evaluate the proposed wavy-attention network architecture, the International Atomic Energy Agency’s Asherah Nuclear Simulator and a false data injection toolbox found in the literature, both implemented in MATLAB/SIMULINK, are utilized. This approach leverages changes in process measurements to identify and classify cyber-attacks on priority signals using the proposed wavy-attention network. Three false data injection attacks are simulated on the simulator’s pressure, temperature, and level sensors to obtain representative process measurements. The wavy-attention network is trained and validated with normal and compromised process variables obtained from the simulator. The performance of the wavy-attention network to discriminate between the reactor states using the test set shows 99% accuracy, as opposed to other baseline models such as vanilla convolution neural networks, long short-term memory networks, and bi-directional long short-term memory networks with 90%, 77%, and 91% accuracy, respectively. An ablation study is also conducted to test the contribution of each component of the proposed architecture. The theoretical framework of the proposed wavy-attention network and its implementation for nuclear reactor digital sensor attack detection are discussed in this paper.

MFMENet: multi-scale features mutual enhancement network for change detection in remote sensing images

ABSTRACT Change detection, an important task in remote sensing image analysis, has been extensively studied in recent years. However, change detection still faces problems such as difficulty in detecting small targets and incomplete edge detection. Furthermore, pseudo changes such as seasonal changes can also lead to many false detections. In response to these challenges, we propose the Multi-scale Features Mutual Enhancement Network (MFMENet), a simple yet efficient network. MFMENet maximizes feature utilization through mutual guidance and supplementation, enhancing the detection capabilities for small targets and edges. First, we use a lightweight feature extraction network to extract features, which mitigates the information loss caused by continuous downsampling of an overly deep network structure. Then, we design a Context Adaptive Interaction Module (CAIM) to realize the complementarity of feature information at different levels. This facilitates shallow features in gaining more semantic information and deep features in acquiring more texture information, thereby enhancing the model’s capability to capture more comprehensive edge features while effectively mitigating interference from pseudo changes. Finally, we introduce a Feature Aggregation Comparison Module (FACM), which uses a combination of aggregation and comparison methods to refine and enhance features. FACM can not only highlight the changed features but also retain more details, improving the model’s detection ability of small targets and edge details. The full utilization of features and effective mutual enhancement of information ensure the improvement of MFMENet’s performance in small target and edge detection. Extensive experiments on three publicly available datasets (LEVIR, DSIFN, and CDD) demonstrate that our approach achieves superior performance with fewer parameters compared to state-of-the-art methods in recent years. In comparison to these baseline methods, our proposed approach achieves improvements of 0.98%, 12.24%, and 2.03% in the IOU metric on the LEVIR, DSIFN, and CDD datasets, respectively, while utilizing only 1.1 M parameters.

Using an Improved Visual Geometry Group Neural Network for Skin Cancer Detection

Over time, deep learning has become more accurate and efficient at dealing with more complex risks. However, training deep neural networks remains a daunting problem and challenges remain, possibly due to several issues, including relevance, generalizability, and training time. Producing specific processors to manipulate artificial networks and reduce task-specific network learning time is natural, but the challenge of fit and generalizability remains undisputed. These networks have been used to detect and diagnose skin cancer, one of the most dangerous types of cancer caused by DNA damage. Automatic recognition of skin cancer is important to help doctors detect the disease in its early stages. In this research, the proposed vgg16 method was used to classify benign and malignant medical images. Image processing was used, where Gaussian noise and salt and pepper noise were added to increase the data, and filters for the arithmetic mean and median were added for each noise. The network was implemented using the Keras interface. Appling suggest model shown that the height accuracy 0.86 for train and accuracy 0.92 for test

BMFNet: Bifurcated multi-modal fusion network for RGB-D salient object detection

Although deep learning-based RGB-D salient object detection methods have achieved impressive results in the recent years, there are still some issues need to be addressed including multi-modal fusion and multi-level aggregation. In this paper, we propose a bifurcated multi-modal fusion network (BMFNet) to address these two issues cooperatively. First, we design a multi-modal feature interaction (MFI) module to fully capture the complementary information between the RGB and depth features by leveraging the channel attention and spatial attention. Second, unlike the widely used layer-by-layer progressive fusion, we adopt a bifurcated fusion strategy for all the multi-level unimodal and cross-modal features to effectively reduce the gaps between features at different levels. For the intra-group feature aggregation, a multi-modal feature fusion (MFF) module is designed to integrate the intra-group multi-modal features to produce a low-level/high-level saliency feature. For the inter-group aggregation, a multi-scale feature learning (MFL) module is introduced to exploit the contextual interactions between different scales to boost fusion performance. Experimental results on five public RGB-D datasets demonstrate the effectiveness and superiority of our proposed network. The code and prediction maps will be available at https://github.com/ZhangQing0329/BMFNet

SCFNet: Lightweight Steel Defect Detection Network Based on Spatial Channel Reorganization and Weighted Jump Fusion

The goal of steel defect detection is to enhance the recognition accuracy and accelerate the detection speed with fewer parameters. However, challenges arise in steel sample detection due to issues such as feature ambiguity, low contrast, and similarity among inter-class features. Moreover, limited computing capability makes it difficult for small and medium-sized enterprises to deploy and utilize networks effectively. Therefore, we propose a novel lightweight steel detection network (SCFNet), which is based on spatial channel reconstruction and deep feature fusion. The network adopts a lightweight and efficient feature extraction module (LEM) for multi-scale feature extraction, enhancing the capability to extract blurry features. Simultaneously, we adopt spatial and channel reconstruction convolution (ScConv) to reconstruct the spatial and channel features of the feature maps, enhancing the spatial localization and semantic representation of defects. Additionally, we adopt the Weighted Bidirectional Feature Pyramid Network (BiFPN) for defect feature fusion, thereby enhancing the capability of the model in detecting low-contrast defects. Finally, we discuss the impact of different data augmentation methods on the model accuracy. Extensive experiments are conducted on the NEU-DET dataset, resulting in a final model achieving an mAP of 81.2%. Remarkably, this model only required 2.01 M parameters and 5.9 GFLOPs of computation. Compared to state-of-the-art object detection algorithms, our approach achieves a higher detection accuracy while requiring fewer computational resources, effectively balancing the model size and detection accuracy.

A dynamic inference network (DI-Net) for online fabric defect detection in smart manufacturing

A dynamic inference network (DI-Net) for online fabric defect detection in smart manufacturing

Application of deep learning to pressure injury staging.

Accurate assessment of pressure injuries (PIs) is necessary for a good outcome. Junior and non-specialist nurses have less experience with PIs and lack clinical practice, and so have difficulty staging them accurately. In this work, a deep learning-based system for PI staging and tissue classification is proposed to help improve its accuracy and efficiency in clinical practice, and save healthcare costs. A total of 1610 cases of PI and their corresponding photographs were collected from clinical practice, and each sample was accurately staged and the tissues labelled by experts for training a Mask Region-based Convolutional Neural Network (Mask R-CNN, Facebook Artificial Intelligence Research, Meta, US) object detection and instance segmentation network. A recognition system was set up to automatically stage and classify the tissues of the remotely uploaded PI photographs. On a test set of 100 samples, the average precision of this model for stage recognition reached 0.603, which exceeded that of the medical personnel involved in the comparative evaluation, including an enterostomal therapist. In this study, the deep learning-based PI staging system achieved the evaluation performance of a nurse with professional training in wound care. This low-cost system could help overcome the difficulty of identifying PIs by junior and non-specialist nurses, and provide valuable auxiliary clinical information.

Domain Feature Decomposition for Efficient Object Detection in Aerial Images

Object detection in UAV aerial images faces domain-adaptive challenges, such as changes in shooting height, viewing angle, and weather. These changes constitute a large number of fine-grained domains that place greater demands on the network’s generalizability. To tackle these challenges, we initially decompose image features into domain-invariant and domain-specific features using practical imaging condition parameters. The composite feature can improve domain generalization and single-domain accuracy compared to the conventional fine-grained domain-detection method. Then, to solve the problem of the overfitting of high-frequency imaging condition parameters, we mixed images from different imaging conditions in a balanced sampling manner as input for the training of the detection network. The data-augmentation method improves the robustness of training and reduces the overfitting of high-frequency imaging parameters. The proposed algorithm is compared with state-of-the-art fine-grained domain detectors on the UAVDT and VisDrone datasets. The results show that it achieves an average detection precision improvement of 5.7 and 2.4, respectively. The airborne experiments validate that the algorithm achieves a 20 Hz processing performance for 720P images on an onboard computer with Nvidia Jetson Xavier NX.

MRMNet: Multi-scale residual multi-branch neural network for object detection

Neural networks in object detection methods are designed to handle the challenging task of multi-scale object detection in computer vision. However, small objects occupy small blocks of pixels, and feature information can easily be overwhelmed, which further leads to certain shortcomings and deficiencies in the network’s multi-scale representation abilities. To alleviate this problem, we propose a novel multi-scale residual multi-branch neural network (MRMNet) for object detection. Particularly, we first propose a multi-scale extension (MSE) module to structurally construct optimal solutions to enhance feature extraction and alleviate information loss, thus effectively improving multi-scale object detection performance. Furthermore, we propose a contextual feature refinement (CFR) module, which draws on the idea of residuals to introduce a multi-residual structure and the feature of partial convolution, which can effectively alleviate the problem of small objects being overwhelmed by conflicting semantic information while avoiding the degradation problem of deep feature networks in some sense. Finally, by combining our own MSE module and integrating it into the existing Modified CSP v6 backbone, we propose a new backbone network, the multi-scale residual (MSRes) network. Experimental results show that our MRMNet and MSRes backbone achieve a competitive detection performance of 43.1% AP on the MS-COCO 2017 dataset.

Adversarial Algorithm Unrolling Network for Interpretable Mechanical Anomaly Detection.

In mechanical anomaly detection, algorithms with higher accuracy, such as those based on artificial neural networks, are frequently constructed as black boxes, resulting in opaque interpretability in architecture and low credibility in results. This article proposes an adversarial algorithm unrolling network (AAU-Net) for interpretable mechanical anomaly detection. AAU-Net is a generative adversarial network (GAN). Its generator, composed of an encoder and a decoder, is mainly produced by algorithm unrolling of a sparse coding model, which is specially designed for feature encoding and decoding of vibration signals. Thus, AAU-Net has a mechanism-driven and interpretable network architecture. In other words, it is ad hoc interpretable. Moreover, a multiscale feature visualization approach for AAU-Net is introduced to verify that meaningful features are encoded by AAU-Net, helping users to trust the detection results. The feature visualization approach enables the results of AAU-Net to be interpretable, i.e., post hoc interpretable. To verify AAU-Net's capability of feature encoding and anomaly detection, we designed and performed simulations and experiments. The results show that AAU-Net can learn signal features that match the dynamic mechanism of the mechanical system. Considering the excellent feature learning ability, unsurprisingly, AAU-Net achieves the best overall anomaly detection performance compared with other algorithms.

Improving Needle Tip Tracking and Detection in Ultrasound-Based Navigation System Using Deep Learning-Enabled Approach.

Ultrasound-guided percutaneous interventions have numerous advantages over traditional techniques. Accurate needle placement in the target anatomy is crucial for successful intervention, and reliable visual information is essential to achieve this. However, previous studies have revealed several challenges, such as the variability in needle echogenicity and the common misalignment of the ultrasound beam and the needle. Advanced techniques have been developed to optimize needle visualization, including hardware-based and image-processing-based methods. This paper proposes a novel strategy of integrating ultrasound-based deep learning approaches into an optical navigation system to enhance needle visualization and improve tip positioning accuracy. Both the tracking and detection algorithms are optimized utilizing optical tracking information. The information is introduced into the tracking network to define the search patch update strategy and form a trajectory reference to correct tracking results. In the detection network, the original image is processed according to the needle insertion position and current position given by the optical localization system to locate a coarse region, and the depth-score criterion is adopted to optimize detection results. Extensive experiments demonstrate that our approach achieves promising tip tracking and detection performance with tip localization errors of 1.11 ± 0.59 mm and 1.17 ± 0.70 mm, respectively. Moreover, we establish a paired dataset consisting of ultrasound images and their corresponding spatial tip coordinates acquired from the optical tracking system and conduct real puncture experiments to verify the effectiveness of the proposed methods. Our approach significantly improves needle visualization and provides physicians with visual guidance for posture adjustment.