Pointer Meter Automatic Recognition (PMAR) under outdoor environment is a challenging task. Due to the variable weather and uneven lighting factors, hand-crafted features or shallow learning techniques have low accuracy in meter recognition. In this paper, a Multitask Cascading Convolutional Neural Network (MC-CNN) is proposed to improve the accuracy of meter recognition under outdoor environment. The proposed MC-CNN used cascaded CNN, including three stages of meter detection, meter cropping and meter reading. Firstly, the YOLOV4 network is used for meter detection to quickly determine the meter location from captured images. In order to accurately cluster pointer meters prior boxes in the YOLOV4 Network, an improved K-means algorithm is presented to further enhance the detection accuracy. Then, the detected meter images are cropped out of the captured images to remove redundant backgrounds. Finally, a meter Reading Network (RNet) based on Adaptive Attention Residual Module (AARM) is proposed for reading meters from cropped images. The proposed AARM not only contains an attention mechanism to focus on essential information and diminish the useless one well, but also extracts information features from meter images adaptively. The experimental results show that the proposed MC-CNN can effectively achieve outdoor meter recognition, with high recognition accuracy and low relative error. The recognition accuracy can reach 92.6%. Compared with the outstanding methods, the average relative error is 2.5655%, reducing by about 3%. What is more, the proposed approach can obtain rich information about the type, limits, units and readings of the pointer meter and can be used when multiple pointer meters exist in one captured image simultaneously. Meanwhile, the proposed approach can significantly improve the accuracy of the recognized readings, and is also robust to the natural environments.
The grey wolf optimizer (GWO) as a new intelligent optimization algorithm has been successfully applied in many fields because of its simple structure, few adjustment parameters and easy implementation. This paper mainly aims at the defects of GWO in path planning application, such as easily falling into local optimization, poor convergence and poor accuracy, and turn point grey wolf optimization (TPGWO) algorithm is proposed. First, the idea of cross-mutation and roulette is used to increase the initial population of GWO and improve the search range. At the same time, the convergence factor function is improved to become a nonlinear update. In the early stage, the search range is expanded, and in the later stage, the convergence speed is increased, while the parameters in the convergence factor function can be adjusted according to the number of obstacles and the map area to change the turning point of the function to improve the convergence speed and accuracy of the algorithm. The turning times and turning angles of the obtained path are added to the fitness function as penalty values to improve the path accuracy. The optimization test is carried out through 16 test functions, and the test results prove the convergence and robustness of TPGWO algorithm. Finally, the TPGWO algorithm is applied to the path planning of patrol robot for simulation experiments. Compared with the GWO algorithm and Particle Swarm Optimization, the simulation results show that the TPGWO algorithm has better convergence, stability and accuracy in the path planning of patrol robot.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.