A functional enhancement on scarred fingerprint using sigmoid filtering

细节 人工智能 计算机科学 乙状窦函数 生物识别 指纹(计算) 滤波器(信号处理) 计算机视觉 图像质量 噪音(视频) 模式识别(心理学) 指纹识别 图像(数学) 人工神经网络
作者
Hoshang Kolivand,Ainul Azura Binti Abdul Hamid,Shiva Asadianfam,Mohd. Shafry Mohd. Rahim
出处
期刊:Neural Computing and Applications [Springer Nature]
卷期号:34 (22): 19973-19994 被引量:1
标识
DOI:10.1007/s00521-022-07520-x
摘要

Abstract Fingerprint has been widely used in biometric applications. Numerous established researches on image enhancement techniques have been done to improve the quality of fingerprint images. However, the production of low-quality images due to the presence of scars remains a challenge in biometrics. The scars damage the fingerprint minutiae information due to broken ridges and they reduce the accuracy of identification. This research developed an image enhancement approach to improve the quality of scarred fingerprint images to generate accurate minutiae extraction. To achieve the aim, the scarred image was improved by removing noise using a new filter, Median Sigmoid (MS), and the corrected ridges were reconstructed using ridges structure enhancement algorithm. This was done to enhance the broken ridges structure. MS filter is a combination of median filter and modified sigmoid function that improves the image contrast and simultaneously removes noise in the fingerprint image. Following that, the filtered image was used in the ridges structure enhancement process. To identify true minutiae, the broken ridges structure in the filtered image needed to be accurately verified. In the ridges structure reconstruction process, an algorithm was enhanced to identify the best value of Sigma parameter ( σ ) used in the Gaussian Low-pass filter to generate a better orientation image. The image is important to reconstruct the corrupted fingerprint ridges structure. The evaluation for the proposed approach used the National Institute of Standards and Technology Special Database 14, and the results showed a 37% improvement of the quality index in comparison to approaches found in related research. The findings of the evaluation showed that the proposed enhancement approach produced a better minutiae extraction result and this is very significant in the field of fingerprint image enhancement.
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
刚刚
Zz完成签到 ,获得积分10
1秒前
橡皮泥发布了新的文献求助30
2秒前
guoguo82完成签到,获得积分10
2秒前
ymX发布了新的文献求助50
2秒前
Neko完成签到,获得积分10
3秒前
liu完成签到,获得积分10
3秒前
拼搏小丸子完成签到 ,获得积分10
4秒前
4秒前
orixero应助谦让的指甲油采纳,获得10
5秒前
Arvilzzz完成签到,获得积分10
7秒前
害羞的裘完成签到 ,获得积分10
7秒前
冷酷的小懒猪完成签到 ,获得积分10
8秒前
9秒前
Hello应助青青草原图图采纳,获得10
9秒前
漂亮的飞莲完成签到,获得积分10
10秒前
科研通AI5应助邱邱采纳,获得30
10秒前
宋枝野完成签到 ,获得积分10
10秒前
11秒前
12秒前
dyc238100完成签到,获得积分10
12秒前
hipig发布了新的文献求助10
14秒前
14秒前
14秒前
风趣的烤鸡完成签到,获得积分20
15秒前
赘婿应助耍酷的白桃采纳,获得10
15秒前
仲夏夜之梦完成签到,获得积分10
16秒前
共享精神应助然大宝采纳,获得10
16秒前
16秒前
xx完成签到 ,获得积分10
16秒前
Arvilzzz发布了新的文献求助10
17秒前
17秒前
达芙发布了新的文献求助10
17秒前
998877剑指完成签到,获得积分10
18秒前
zq1992nl完成签到,获得积分10
21秒前
搜集达人应助大白杨采纳,获得10
21秒前
酷波er应助体贴精灵采纳,获得10
21秒前
华仔应助石思炜采纳,获得10
21秒前
CC发布了新的文献求助30
22秒前
任世界灯火阑珊完成签到,获得积分10
23秒前
高分求助中
Continuum Thermodynamics and Material Modelling 3000
Production Logging: Theoretical and Interpretive Elements 2700
Les Mantodea de Guyane Insecta, Polyneoptera 1000
Structural Load Modelling and Combination for Performance and Safety Evaluation 1000
Conference Record, IAS Annual Meeting 1977 820
England and the Discovery of America, 1481-1620 600
電気学会論文誌D(産業応用部門誌), 141 巻, 11 号 510
热门求助领域 (近24小时)
化学 材料科学 生物 医学 工程类 有机化学 生物化学 物理 纳米技术 计算机科学 内科学 化学工程 复合材料 基因 遗传学 物理化学 催化作用 量子力学 光电子学 冶金
热门帖子
关注 科研通微信公众号,转发送积分 3572102
求助须知:如何正确求助?哪些是违规求助? 3142380
关于积分的说明 9447398
捐赠科研通 2843806
什么是DOI,文献DOI怎么找? 1563098
邀请新用户注册赠送积分活动 731575
科研通“疑难数据库(出版商)”最低求助积分说明 718603