计算机与数字工程2019，Vol.47Issue(12)：3142-3148,7.DOI:10. 3969/j. issn. 1672-9722. 2019. 12. 038
基于多尺度空洞卷积自编码神经网络的森林烟火监测
Forest Fire Detection Based on Multi-Scale Dilated Convolution Auto-Encoding Neural Network
摘要
Abstract
Large range of forest fires cause the large size of effective feature in fire images,which are difficult to be learned ef?fectively by traditional convolution networks. In addition,because fire smoke and fog are similar,it is easily to recognize error. Aim?ing at these problems,a new deep network based on weighted multi-scale dilated convolution auto-encoder(MSDCAE)is pro?posed. Different sizes of receptive field features are obtained through different dilated convolution kernels and concatenated output to optimize feature learning. Based on softmaxwithloss,improved loss function(ISWL)is designed to improve the classification perfor?mance of similar images such as fire smoke and fog. Experiments verify the effectiveness of MSDCAE auto-encoder and ISWL loss function. The results prove that the new method is superior to the ordinary deep network algorithm in the image recognition of forest fires.关键词
烟火监测/空洞卷积/特征提取/神经网络Key words
fire smoke detect/dilated convolution/extract features/neural network分类
农业科技引用本文复制引用
冯嘉良,朱定局,廖丽华..基于多尺度空洞卷积自编码神经网络的森林烟火监测[J].计算机与数字工程,2019,47(12):3142-3148,7.基金项目
国家社会科学基金重大项目(编号:14ZDB101) （编号:14ZDB101）
国家自然基金项目(编号:61105133) （编号:61105133）
广东省联合培养研究生示范基地(编号:粤教研函[2016]39号) （编号:粤教研函[2016]39号）
广东省新工科研究与实践项目(编号:粤教高函[2017]118号) （编号:粤教高函[2017]118号）
广东省高等教育教学研究和改革项目(编号:粤教高函[2016]236号)资助. （编号:粤教高函[2016]236号）
