Detection Algorithm of Laboratory Personnel Irregularities Based on Improved YOLOv7

Due to the complex environment of the university laboratory,personnel flow intensive,personnel irregular behavior is easy to cause security risks.Monitoring using mainstream detection algorithms suffers from low detection accuracy and slow speed.Therefore,the current management of personnel behavior mainly relies on institutional constraints,education and training,on-site supervision,etc.,which is time-consuming and ineffective.Given the above situation,this paper proposes an improved You Only Look Once version 7(YOLOv7)to achieve the purpose of quickly detecting irregular behaviors of laboratory personnel while ensuring high detection accuracy.First,to better capture the shape features of the target,deformable convolutional networks(DCN)is used in the backbone part of the model to replace the traditional convolution to improve the detection accuracy and speed.Second,to enhance the extraction of important features and suppress useless features,this paper proposes a new convolutional block attention module_efficient channel attention(CBAM_E)for embedding the neck network to improve the model’s ability to extract features from complex scenes.Finally,to reduce the influence of angle factor and bounding box regression accuracy,this paper proposes a newα-SCYLLA intersection over union(α-SIoU)instead of the complete intersection over union(CIoU),which improves the regression accuracy while increasing the convergence speed.Comparison experiments on public and homemade datasets show that the improved algorithm outperforms the original algorithm in all evaluation indexes,with an increase of 2.92%in the precision rate,4.14%in the recall rate,0.0356 in the weighted harmonic mean,3.60%in the mAP@0.5 value,and a reduction in the number of parameters and complexity.Compared with the mainstream algorithm,the improved algorithm has higher detection accuracy,faster convergence speed,and better actual recognition effect,indicating the effectiveness of the improved algorithm in this paper and its potential for practical application in laboratory scenarios.

Feature deformation network with multi-range feature enhancement for agricultural machinery operation mode identification

Utilizing the spatiotemporal features contained in extensive trajectory data for identifying operation modes of agricultural machinery is an important basis task for subsequent agricultural machinery trajectory research.In the present study,to effectively identify agricultural machinery operation mode,a feature deformation network with multi-range feature enhancement was proposed.First,a multi-range feature enhancement module was developed to fully explore the feature distribution of agricultural machinery trajectory data.Second,to further enrich the representation of trajectories,a feature deformation module was proposed that can map trajectory points to high-dimensional space to form feature maps.Then,EfficientNet-B0 was used to extract features of different scales and depths from the feature map,select features highly relevant to the results,and finally accurately predict the mode of each trajectory point.To validate the effectiveness of the proposed method,experiments were conducted to compare the results with those of other methods on a dataset of real agricultural trajectories.On the corn and wheat harvester trajectory datasets,the model achieved accuracies of 96.88%and 96.68%,as well as F1 scores of 93.54%and 94.19%,exhibiting improvements of 8.35%and 9.08%in accuracy and 20.99%and 20.04%in F1 score compared with the current state-of-the-art method.

A Remote Sensing Image Semantic Segmentation Method by Combining Deformable Convolution with Conditional Random Fields

Currently,deep convolutional neural networks have made great progress in the field of semantic segmentation.Because of the fixed convolution kernel geometry,standard convolution neural networks have been limited the ability to simulate geometric transformations.Therefore,a deformable convolution is introduced to enhance the adaptability of convolutional networks to spatial transformation.Considering that the deep convolutional neural networks cannot adequately segment the local objects at the output layer due to using the pooling layers in neural network architecture.To overcome this shortcoming,the rough prediction segmentation results of the neural network output layer will be processed by fully connected conditional random fields to improve the ability of image segmentation.The proposed method can easily be trained by end-to-end using standard backpropagation algorithms.Finally,the proposed method is tested on the ISPRS dataset.The results show that the proposed method can effectively overcome the influence of the complex structure of the segmentation object and obtain state-of-the-art accuracy on the ISPRS Vaihingen 2D semantic labeling dataset.

Gravity variation in the Tibet area before the Nepal Ms8.1 earthquake

This research utilized two periods of gravity monitoring results from 2010 to 2013 from the Continental Tectonics Environmental Monitoring Network of China, analyzed the corre- lation between gravity variation in the Tibet area and the Nepal Ms8.1 earthquake, and investigated the gravity variation mechanism in combination with the crust vertical movement and horizontal strain field observed by Global Positioning System （GPS）. The research results indicated that （1） the gravity variation exhibited apparent characteristics of a positive anomaly and high gradient zone in the Himalayan frontier. This observation is consistent with the existing recognition of the gravity anomaly and occurrence regularity of a strong earthquake; （2） the gravity variation exhibited apparent consistence with the spacious distribution of the vertical movement and the horizontal deformation field in that area. The crustal vertical movement was not the direct cause leading to the gravity vari- ation. It is assumed that the crust stress-strain accumulation in the Qinghai-Tibetan Plateau and its adjacent areas is the important factor that resulted in the variation of gravity.

DSD-MatchingNet:Deformable sparse-to-dense feature matching for learning accurate correspondences

Background Exploring correspondences across multiview images is the basis of various computer vision tasks.However,most existing methods have limited accuracy under challenging conditions.Method To learn more robust and accurate correspondences,we propose DSD-MatchingNet for local feature matching in this study.First,we develop a deformable feature extraction module to obtain multilevel feature maps,which harvest contextual information from dynamic receptive fields.The dynamic receptive fields provided by the deformable convolution network ensure that our method obtains dense and robust correspondence.Second,we utilize sparse-to-dense matching with symmetry of correspondence to implement accurate pixel-level matching,which enables our method to produce more accurate correspondences.Result Experiments show that our proposed DSD-MatchingNet achieves a better performance on the image matching benchmark,as well as on the visual localization benchmark.Specifically,our method achieved 91.3%mean matching accuracy on the HPatches dataset and 99.3%visual localization recalls on the Aachen Day-Night dataset.