亲爱的研友该休息了!由于当前在线用户较少,发布求助请尽量完整地填写文献信息,科研通机器人24小时在线,伴您度过漫漫科研夜!身体可是革命的本钱,早点休息,好梦!

Smoke-Aware Global-Interactive Non-Local Network for Smoke Semantic Segmentation

计算机科学 分割 稳健性(进化) 人工智能 语义学(计算机科学) 可扩展性 模式识别(心理学) 生物化学 化学 数据库 基因 程序设计语言
作者
Lin Zhang,Jing Wu,Feiniu Yuan,Yuming Fang
出处
期刊:IEEE transactions on image processing [Institute of Electrical and Electronics Engineers]
卷期号:33: 1175-1187 被引量:14
标识
DOI:10.1109/tip.2024.3359816
摘要

Compared with other objects, smoke semantic segmentation (SSS) is more difficult and challenging due to some special characteristics of smoke, such as non-rigid, translucency, variable mode and so on. To achieve accurate positioning of smoke in real complex scenes and promote the development of intelligent fire detection, we propose a Smoke-Aware Global-Interactive Non-local Network (SAGINN) for SSS, which harness the power of both convolution and transformer to capture local and global information simultaneously. Non-local is a powerful means for modeling long-range context dependencies, however, friendliness to single-scale low-resolution features limits its potential to produce high-quality representations. Therefore, we propose a Global-Interactive Non-local (GINL) module, leveraging global interaction between multi-scale key information to improve the robustness of feature representations. To solve the interference of smoke-like objects, a Pyramid High-level Semantic Aggregation (PHSA) module is designed, where the learned high-level category semantics from classification aids model by providing additional guidance to correct the wrong information in segmentation representations at the image level and alleviate the inter-class similarity problem. Besides, we further propose a novel loss function, termed Smoke-aware loss (SAL), by assigning different weights to different objects contingent on their importance. We evaluate our SAGINN on extensive synthetic and real data to verify its generalization ability. Experimental results show that SAGINN achieves 83% average mIoU on the three testing datasets (83.33%, 82.72% and 82.94%) of SYN70K with an accuracy improvement of about 0.5%, 0.002 mMse and 0.805 F β on SMOKE5K, which can obtain more accurate location and finer boundaries of smoke, achieving satisfactory results on smoke-like objects.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
康子浩发布了新的文献求助10
1秒前
贪玩的夏真完成签到,获得积分10
2秒前
雨落瑾年完成签到,获得积分0
12秒前
lin完成签到,获得积分10
12秒前
Glufo完成签到,获得积分10
14秒前
俏皮的灵完成签到,获得积分10
14秒前
16秒前
臭学医的完成签到 ,获得积分10
17秒前
大溺完成签到 ,获得积分10
20秒前
传奇3的应助被DRJ采纳,获得10
24秒前
思源的应助被雪山大地采纳,获得10
24秒前
26秒前
Aixx完成签到 ,获得积分10
30秒前
可靠友梅完成签到,获得积分10
31秒前
YU完成签到,获得积分10
35秒前
Leowo完成签到,获得积分10
35秒前
和谐以冬完成签到,获得积分10
37秒前
39秒前
深情安青的应助被Qiaoguliang采纳,获得10
40秒前
40秒前
Moooi完成签到,获得积分10
43秒前
44秒前
莫里亚蒂完成签到,获得积分10
47秒前
Charlie完成签到 ,获得积分10
48秒前
Yu完成签到 ,获得积分10
49秒前
可爱的函函的应助被DRJ采纳,获得10
50秒前
51秒前
54秒前
55秒前
莫里亚蒂完成签到,获得积分10
56秒前
56秒前
Ch发布了新的文献求助10
58秒前
NexusExplorer的应助被wstki采纳,获得10
58秒前
哈哈哈哈发布了新的文献求助10
59秒前
1分钟前
1分钟前
高分子聚合绿豆汤完成签到 ,获得积分10
1分钟前
勤奋柜子的应助被雪山大地采纳,获得10
1分钟前
Ahloe发布了新的文献求助10
1分钟前
JohnYoung发布了新的文献求助10
1分钟前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
Hidden Empire of Finance: How Wall Street Profits from Our Cities and Fuels Global Inequality 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8165776
求助须知:如何正确求助?哪些是违规求助? 9559273
关于积分的说明 21749731
捐赠科研通 7756580
什么是DOI,文献DOI怎么找? 3429034
关于科研通互助平台的介绍 2553319
邀请新用户注册赠送积分活动 2473606