In order to solve the poor performance of real-time semantic segmentation of road conditions in video images due to insufficient light and motion blur when vehicles are driving at night, This study proposes a scheme: a fuzzy information complementation strategy based on generative models and a network that fuses different intermediate layer outputs to complement spatial semantics with also embeds irregular convolutional attention modules for fine extraction of motion target boundaries. First, DeblurGan is used to generate information to fix the lost semantics in the original image due to blurring; then, the outputs of different intermediate layers in the backbone network are taken out, assigned different weight scaling factors and fused; finally, by comparing the performance of different attention mechanisms, the irregular convolutional attention with the best effect is selected. The scheme achieves Global Accuracy:89.1% Mean IOU:94.2% on the night driving dataset of this experiment, which exceeds the best performance of DeepLabv3 by 1.3% and 7.2%, and achieves Accuracy:83.0% on the small volume label (Moveable), which is underperformed by all control models. The experimental results demonstrate that the solution can effectively cope with various problems faced by night driving and enhance the model's perception and analysis of driving road conditions. The results of the study provide a technical reference for the semantic segmentation problem of vehicles driving in the nighttime environment.
In order to solve the poor performance of real-time semantic segmentation of night road conditions in video images due to insufficient light and motion blur, this study proposes a scheme: a fuzzy information complementation strategy based on generative models and a network that fuses different intermediate layer outputs to complement spatial semantics which also embeds irregular convolutional attention modules for fine extraction of motion target boundaries. First, DeblurGan is used to generate information to fix the lost semantics in the original image; then, the outputs of different intermediate layers are taken out, assigned different weight scaling factors, and fused; finally, the irregular convolutional attention with the best effect is selected. The scheme achieves Global Accuracy of 89.1% Mean and IOU 94.2% on the night driving dataset of this experiment, which exceeds the best performance of DeepLabv3 by 1.3 and 7.2%, and achieves an Accuracy of 83.0% on the small volume label (Moveable). The experimental results demonstrate that the solution can effectively cope with various problems faced by night driving and enhance the model's perception. It also provides a technical reference for the semantic segmentation problem of vehicles driving in the nighttime environment.
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.