Research on Lightweight Method of Insulator Target Detection Based on Improved SSD

Sensors, Journal Year: 2024, Volume and Issue: 24(18), P. 5910 - 5910

Published: Sept. 12, 2024

Aiming at the problems of a large volume, slow processing speed, and difficult deployment in edge terminal, this paper proposes lightweight insulator detection algorithm based on an improved SSD. Firstly, original feature extraction network VGG-16 is replaced by Ghost Module to initially achieve model. A Feature Pyramid structure Network (FPN+PAN) are integrated into Neck part Simplified Spatial Pooling Fast (SimSPPF) module introduced realize integration local features global features. Secondly, multiple Channel Squeeze-and-Excitation (scSE) attention mechanisms make model pay more channels containing important information. The six heads reduced four improve inference speed network. In order recognition performance occluded overlapping targets, DIoU-NMS was used replace non-maximum suppression (NMS). Furthermore, channel pruning strategy reduce unimportant weight matrix model, knowledge distillation fine-adjust after pruning, so as ensure accuracy. experimental results show that parameter number proposed from 26.15 M 0.61 M, computational load 118.95 G 1.49 G, mAP increased 96.8% 98%. Compared with other models, not only guarantees accuracy algorithm, but also greatly reduces which provides support for realization visible light target intelligence.

Language: Английский

Accurate Prediction of 327 Rice Variety Growth Period Based on Unmanned Aerial Vehicle Multispectral Remote Sensing

Drones, Journal Year: 2024, Volume and Issue: 8(11), P. 665 - 665

Published: Nov. 10, 2024

Most rice growth stage predictions are currently based on a few varieties for prediction method studies, primarily using linear regression, machine learning, and other methods to build models that tend have poor generalization ability, low accuracy, face various challenges. In this study, multispectral images of at stages were captured an unmanned aerial vehicle, single-plant silhouettes identified 327 by establishing deep-learning algorithm. A was established the normalized vegetation index combined with cubic polynomial regression equations simulate their changes, it first proposed different inferred analyzing difference rate. Overall, contour recognition model showed good ability varieties, most accuracies in range 0.75–0.93. The accuracy recognizing also some variation, root mean square error between 0.506 3.373 days, relative 2.555% 14.660%, Bias between1.126 2.358 0.787% 9.397%; therefore, can be used effectively improve periods rice.

Language: Английский