An Improved Iris Segmentation Technique Using Circular Hough Transform

Okokpujie, Kennedy; Noma‐Osaghae, Etinosa; John, Samuel; Ajulibe, Akachukwu

doi:10.1007/978-981-10-6454-8_26

Cited by 31 publications

(12 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1) Iris segmentation algorithms based on image gradients: Okokpujie et al [4] proposed an iris segmentation algorithm that replaces integro-differential operators with the Hough transform, which increases the segmentation accuracy of the algorithm. Umer et al [5] proposed a restricted Hough transform algorithm, which reduces the parameter search space of the traditional Hough transform algorithms, and the algorithm spends less time during the segmentation process and obtains higher accuracy.…”

Section: Related Workmentioning

confidence: 99%

An Adaptive CNNs Technology for Robust Iris Segmentation

et al. 2019

View full text Add to dashboard Cite

Iris segmentation algorithms are of great significance in complete iris recognition systems, and directly affect the iris verification and recognition results. However, the conventional iris segmentation algorithms have poor adaptability and are not sufficiently robust when applied to noisy iris databases captured under unconstrained conditions. In addition, there are currently no large iris databases; thus, the iris segmentation algorithms cannot maximize the benefits of convolutional neural networks (CNNs). The main work of this paper is as follows: first, we propose an architecture based on CNNs combined with dense blocks for iris segmentation, referred to as a dense-fully convolutional network (DFCN), and adopt some popular optimizer methods, such as batch normalization (BN) and dropout. Second, because the public ground-truth masks of the CASIA-Interval-v4 and IITD iris databases do not include the labeled eyelash regions, we label these regions that occlude the iris regions using the Labelme software package. Finally, the promising results of experiments based on the CASIA-Interval-v4, IITD, and UBIRIS.V2 iris databases captured under different conditions reveal that the iris segmentation network proposed in this paper outperforms all of the conventional and most of the CNN-based iris segmentation algorithms with which we compared our algorithm's results in terms of various metrics, including the accuracy, precision, recall, f1 score, and nice1 and nice2 error scores, reflecting the robustness of our proposed network. INDEX TERMS CNNs, dense block, dense-fully convolutional network, iris segmentation.

show abstract

Section: Related Workmentioning

confidence: 99%

An Adaptive CNNs Technology for Robust Iris Segmentation

et al. 2019

View full text Add to dashboard Cite

show abstract

“…With a database of nine (9) individuals and ten (10) images for each individual, the designed system showed accuracies as high as 90% and as low as 70%. The accuracy of the designed system can be improved.…”

Section: Discussionmentioning

confidence: 99%

Development of a facial recognition system with email identification message relay mechanism

Okokpujie

Noma‐Osaghae

John

et al. 2017

2017 International Conference on Computing Networking and Informatics (ICCNI)

Self Cite

View full text Add to dashboard Cite

Abstract-Attendance records play a vital role in the educational sector. It is so vital that students are not allowed to sit for examinations if they do not meet the class attendance benchmark. But students, instead of making sure they attend classes regularly, devise cunny ways of committing attendance fraud. This unpleasant trend has made it necessary to develop systems that can take accurate class attendance records and minimize fraud. The use of biometrics to develop attendance taking systems is becoming quite popular. One of such biometrics is The Face. In this paper, a facial recognition algorithm known as Fisherfaces or Fisher Discriminant Analysis (FDA) which is not sensitive to substantial variation in facial look and illumination is used to develop the facial recognition attendance taking system. The system implemented has a training database of Ten (10) students. Ten (10) facial images of each student are taken with different composures, looks and under different levels of illumination. Tests on nine (9) students in the database yielded accuracies of as low as 70% and as high as 90%. This validates the proof that the more the number of training facial image in the database, the higher the accuracy of Fisherfaces approach. The simple mail transfer protocol (SMTP) was interfaced with the database to send identification messages (name of student identified with time and date of identification) to the email address of the administrator (in this case the lecturer) in realtime to effectively monitor the attendance. The result was found capable of eliminating attendance fraud.

show abstract

“…In addition, Koon et al took advantage of parabolic Hough transform to detect the upper and lower eyelids using parabolic arcs, as depicted in [12]. Moreover, many research work have used Hough transform algorithm as platform for segmenting the iris image to round circle as posited in [11], [9] [13], [14] [1] [13] [15].…”

Section: A Segmentationmentioning

confidence: 99%

“…Hence, with this reason it is necessary to set a proper threshold when using Hamming distance. This promotes more accurate in identification [14]. In addition, much research work used Hamming distance for measuring the similarity between the two iris templates as posited in [11] III.…”

Section: Matchingmentioning

confidence: 99%

A Review of Iris Recognition Algorithms

Ghali

Jamel

Mohamad

et al. 2017

JOIV : Int. J. Inform. Visualization

View full text Add to dashboard Cite

Abstract-With the prominent needs for security and reliable mode of identification in biometric system. Iris recognition has become reliable method for personal identification nowadays. The system has been used for years in many commercial and government applications that allow access control in places such as office, laboratory, armoury, automated teller machines (ATMs), and border control in airport. The aim of the paper is to review iris recognition algorithms. Iris recognition system consists of four main stages which are segmentation, normalization, feature extraction and matching. Based on the findings, the Hough transform, rubber sheet model, wavelet, Gabor filter, and hamming distance are the most common used algorithms in iris recognition stages. This shows that, the algorithms have the potential and capability to enhanced iris recognition system.

show abstract

An Improved Iris Segmentation Technique Using Circular Hough Transform

Cited by 31 publications

References 3 publications

An Adaptive CNNs Technology for Robust Iris Segmentation

An Adaptive CNNs Technology for Robust Iris Segmentation

Development of a facial recognition system with email identification message relay mechanism

A Review of Iris Recognition Algorithms

Contact Info

Product

Resources

About