Pixel‐level thin crack detection on road surface using convolutional neural network for severely imbalanced data

像素 卷积神经网络 增采样 背景(考古学) 计算机科学 卷积(计算机科学) 人工智能 路面 网(多面体) 人工神经网络 模式识别(心理学) 图像(数学) 计算机视觉 数学 材料科学 地质学 几何学 复合材料 古生物学
作者
Thitirat Siriborvornratanakul
出处
期刊:Computer-aided Civil and Infrastructure Engineering [Wiley]
卷期号:38 (16): 2300-2316 被引量:29
标识
DOI:10.1111/mice.13010
摘要

Abstract Because roads are the major backbone of the transportation network, research about crack detection on road surfaces has been popular in computer and infrastructure engineering. When training a convolutional neural network (CNN) for pixel‐level road crack detection, three common challenges include (1) the data are severely imbalanced, (2) crack pixels can be easily confused with normal road texture and other visual noises, and (3) there are many unexplainable characteristics regarding the CNN itself. When it comes to very fine and thin cracks, these challenges are exaggerated and a new challenge is introduced, as there can be a discrepancy between the actual width and the annotated width of a crack. To tackle all these challenges of thin crack detection, this paper proposes a new variant of CNN named ThinCrack U‐Net, designed to provide thin results upon pixel‐level crack detection on road surfaces. The main contribution is to demystify how pixel‐level thin crack detection results are affected by different loss functions as well as various combinations of the U‐Net components. The experimental results show that ThinCrack U‐Net yields a significant performance boost in CrackTree260, from 65.71% to 94.48% F‐measure, compared to the existing variant of U‐Net previously proposed in the context of pixel‐level thin crack detection. Finally, this paper locates the source of undesirable result thickness and solves it with the balanced usage of downsampling/upsampling layers and atrous convolution. Unlike suggested by previous works, different loss functions show no significant impact on ThinCrack U‐Net, whereas normalization layers are proved crucial in pixel‐level thin crack detection.
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
小星星发布了新的文献求助10
1秒前
小马甲应助Nomb1采纳,获得10
2秒前
FashionBoy应助章慕思采纳,获得10
2秒前
BELIEVE完成签到 ,获得积分10
2秒前
ywayw完成签到,获得积分10
2秒前
英姑应助浅斟低唱采纳,获得10
3秒前
Hello应助guo采纳,获得10
4秒前
5秒前
6秒前
Akim应助白茶泡泡球采纳,获得10
8秒前
qcl发布了新的文献求助10
9秒前
9秒前
快去爬山完成签到 ,获得积分10
11秒前
吕小布完成签到,获得积分10
12秒前
12秒前
Sjingjia完成签到,获得积分10
12秒前
Melt发布了新的文献求助10
12秒前
魔女完成签到 ,获得积分10
13秒前
HongJiang完成签到,获得积分10
14秒前
科研通AI5应助Ki_Ayasato采纳,获得10
15秒前
科研通AI2S应助吕小布采纳,获得10
15秒前
penghuiye完成签到,获得积分10
16秒前
无限秋天发布了新的文献求助10
16秒前
芒果布丁完成签到 ,获得积分10
17秒前
17秒前
小星星完成签到,获得积分10
19秒前
20秒前
五十一完成签到 ,获得积分10
20秒前
温婉的松鼠完成签到 ,获得积分10
22秒前
qcl完成签到,获得积分10
23秒前
小子完成签到,获得积分20
23秒前
被文献折磨疯了完成签到,获得积分10
23秒前
晨露完成签到 ,获得积分10
23秒前
Melt完成签到,获得积分10
23秒前
酷波er应助Felixsun采纳,获得10
24秒前
张zi完成签到,获得积分10
25秒前
28秒前
美丽沛春完成签到,获得积分10
29秒前
淡然善斓完成签到,获得积分10
30秒前
含蓄的灵煌完成签到,获得积分10
31秒前
高分求助中
Production Logging: Theoretical and Interpretive Elements 2700
Neuromuscular and Electrodiagnostic Medicine Board Review 1000
こんなに痛いのにどうして「なんでもない」と医者にいわれてしまうのでしょうか 510
The First Nuclear Era: The Life and Times of a Technological Fixer 500
Unusual formation of 4-diazo-3-nitriminopyrazoles upon acid nitration of pyrazolo[3,4-d][1,2,3]triazoles 500
岡本唐貴自伝的回想画集 500
Distinct Aggregation Behaviors and Rheological Responses of Two Terminally Functionalized Polyisoprenes with Different Quadruple Hydrogen Bonding Motifs 450
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 物理 生物化学 纳米技术 计算机科学 化学工程 内科学 复合材料 物理化学 电极 遗传学 量子力学 基因 冶金 催化作用
热门帖子
关注 科研通微信公众号,转发送积分 3671635
求助须知:如何正确求助?哪些是违规求助? 3228335
关于积分的说明 9779690
捐赠科研通 2938645
什么是DOI,文献DOI怎么找? 1610206
邀请新用户注册赠送积分活动 760547
科研通“疑难数据库(出版商)”最低求助积分说明 736093