This paper proposes an improved steel surface defect detection algorithm YOLOv5-TBC based on YOLOv5s to address the issues of insufficient defect feature extraction capability and inadequate feature fusion in steel surface defect detection tasks.Firstly,a triplet attention mechanism is introduced into the core feature extraction module of the Backbone to enhance its adaptability and representational ability.Secondly,a weighted Bi-directional Feature Pyramid Network(BiFPN)is introduced to improve feature fusion,and a CBAM attention mechanism is introduced to optimize the model's detection capability for small-scale targets.Finally,a lightweight upsampling operator CARAFE is added to enlarge the model's receptive field,further improving the detection performance for targets of different sizes.Experimental results demonstrate that the improved YOLOv5s model achieves a precision(P)of 74.0%and a mean Average Precision(mAP)of 76.6%on the NEU-DET dataset,which is an increase of 5.8%in P and 3.0%in mAP compared to YOLOv5s,and an increase of 3.8%in P and 2.1%in mAP compared to YOLOv7,proving that the network model has better detection performance.
A new prediction framework is proposed to improve short-term power load forecasting accuracy. The framework is based on particle swarm optimization (PSO)-variational mode decomposition (VMD) combined with a time convolution network (TCN) embedded attention mechanism (Attention). The framework follows a two-step process. In the first step, PSO is applied to optimize the VMD decomposition method. The original electricity load sequence is decomposed, and the fitness function uses sample entropy to describe the complexity of the time series. The decomposed sub-sequences are combined with relevant features, such as meteorological data, to form the input sequence of the prediction model. In the second step, TCN is selected as the prediction model, and it is embedded with an attention mechanism to improve prediction accuracy. The above input sequence is fed to the model to obtain the PSO-VMD-TCN-Attention prediction framework. Load datasets and various prediction models validate the PSO-optimized VMD decomposition method and the TCN-Attention prediction model. Simulation results demonstrate that the PSO-optimized VMD decomposition method enhances the model’s prediction accuracy, and the TCN-Attention prediction model outperforms other prediction models in terms of prediction accuracy and ability.
In recent years, artificial intelligence has been applied to 3D COVID-19 medical image diagnosis, which reduces detection costs and missed diagnosis rates with higher predictive accuracy, and diagnostic efficiency. However, the limited size and low quality of clinical 3D medical image samples have hindered the segmentation performance of 3D models. Therefore, we propose a 3D medical image segmentation model based on semi-supervised learning using co-training. Multi-view and multi-modal images are generated using spatial flipping and windowing techniques to enhance the spatial diversity of 3D image samples. A pseudo label generation module based on confidence-weights is employed to generate reliable pseudo labels for non-annotated data, thereby increasing the sample size and reducing overfitting. The proposed approach utilizes a three-stage training process: firstly, training a single network based on annotated data; secondly, incorporating non-annotated data to train a dual-modal network and generate pseudo labels; finally, jointly training six models in three dimensions using both annotated and pseudo labels generated from multi-view and multi-modal images, aiming to enhance segmentation accuracy and generalization performance. Additionally, a consistency regularization loss is applied to reduce noises and accelerate convergence of the training. Moreover, a heatmap visualization method is employed to focus on the attention of features at each stage of training, providing effective reference for clinical diagnosis. Experiments were conducted on an open dataset of 3D COVID-19 CT samples and a non-annotated dataset from TCIA, including 771 NIFTI-format CT images from 661 COVID-19 patients. The results of 5-fold cross-validation show that the proposed model achieves a segmentation accuracy of Dice=73.30%, ASD=10.633, Sensitivity=63.00%, and Specificity=99.60%. Compared to various typical semi-supervised learning 3D segmentation models, it demonstrates better segmentation accuracy and generalization performance.
Timely and accurate air quality prediction data is very important for environmental management,especially,during the period of heavy air pollution.The prediction data can provide data support for the decision-making of the government's ecological environment management departments to cope with the pollution situation and accurately allocate social resources.The artificial intelligence based deep learning model AirNet6 proposed in this article can give consideration to both accuracy and real-time performance to achieve 7-day or longer air quality prediction for ozone,sulfur dioxide,carbon monoxide and other factors.Unlike traditional chemical model calculations,this model base on Spatio-Temporal Graph Convolutional Networks(STGCN),which captures the laws of historical monitoring data,weather prediction data,social activities and other data,and completes the prediction of more than one hundred points for the next 168 hours in two minutes.Experiments show that the AirNet6 model has made significant progress in speed,energy efficiency,and accuracy compared to traditional chemical models and time series AI models.
Multimodal image registration is the fundamental technique for scene analysis with series remote sensing images of different spectrum region. Due to the highly nonlinear radiometric relationship, it is quite challenging to find common features between images of different modal types. This paper resorts to the deep neural network, and tries to learn descriptors for multimodal image patch matching, which is the key issue of image registration. A Siamese fully convolutional network is set up and trained with a novel loss function, which adopts the strategy of maximizing the feature distance between positive and hard negative samples. The two branches of the Siamese network are connected by the convolutional operation, resulting in the similarity score between the two input image patches. The similarity score value is used, not only for correspondence point location, but also for outlier identification. A generalized workflow for deep feature based multimodal RS image registration is constructed, including the training data curation, candidate feature point generation, and outlier removal. The proposed network is tested on a variety of optical, near infrared, thermal infrared, SAR, and map images. Experiment results verify the superiority over other state-of-the-art approaches.
