Deep Sclera Segmentation and Recognition

Rot, Peter; Vitek, Matej; Grm, Klemen; Emeršič, Žiga; Peer, Peter

doi:10.1007/978-3-030-27731-4_13

Cited by 44 publications

(28 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Following standard methodology [ 32 , 38 , 39 ], the following error rates commonly used for verification experiments are adopted to evaluate the performance of the tested models: Area Under the Curve (AUC): AUC represents a performance metric that measures the overall performance of a learned binary model and is typically computed from a standard Receiver Operating Characteristic (ROC) curve. This metric is widely used to assess performance of biometric systems operating in verification mode.…”

Section: Methodsmentioning

confidence: 99%

Learning to Combine Local and Global Image Information for Contactless Palmprint Recognition

Stoimchev

Ivanovska

2021

Sensors

Self Cite

View full text Add to dashboard Cite

In the past few years, there has been a leap from traditional palmprint recognition methodologies, which use handcrafted features, to deep-learning approaches that are able to automatically learn feature representations from the input data. However, the information that is extracted from such deep-learning models typically corresponds to the global image appearance, where only the most discriminative cues from the input image are considered. This characteristic is especially problematic when data is acquired in unconstrained settings, as in the case of contactless palmprint recognition systems, where visual artifacts caused by elastic deformations of the palmar surface are typically present in spatially local parts of the captured images. In this study we address the problem of elastic deformations by introducing a new approach to contactless palmprint recognition based on a novel CNN model, designed as a two-path architecture, where one path processes the input in a holistic manner, while the second path extracts local information from smaller image patches sampled from the input image. As elastic deformations can be assumed to most significantly affect the global appearance, while having a lesser impact on spatially local image areas, the local processing path addresses the issues related to elastic deformations thereby supplementing the information from the global processing path. The model is trained with a learning objective that combines the Additive Angular Margin (ArcFace) Loss and the well-known center loss. By using the proposed model design, the discriminative power of the learned image representation is significantly enhanced compared to standard holistic models, which, as we show in the experimental section, leads to state-of-the-art performance for contactless palmprint recognition. Our approach is tested on two publicly available contactless palmprint datasets—namely, IITD and CASIA—and is demonstrated to perform favorably against state-of-the-art methods from the literature. The source code for the proposed model is made publicly available.

show abstract

Section: Methodsmentioning

confidence: 99%

Learning to Combine Local and Global Image Information for Contactless Palmprint Recognition

Stoimchev

Ivanovska

2021

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…The jaundice is affected by the white part in the eye, so it should be segmented using the min-max sclera segmentation. Sclera segmentation is a very important process [15]. Min-max color space algorithm works well by converting the image into the color spaces as RGB split them into their respective channels and then find the min/max value using the minMaxLoc function by minimum and maximum value, the white part is sclera part is segmented from the eye image.…”

Section: Segmentationmentioning

confidence: 99%

Early Detection of Pancreatic Cancer Using Jaundiced Eye Images

Roy¹,

Mala²

2022

Computer Systems Science and Engineering

View full text Add to dashboard Cite

Pancreatic cancer is one of the deadliest cancers, with less than 9% survival rates. Pancreatic Ductal Adeno Carcinoma (PDAC) is common with the general public affecting most people older than 45. Early detection of PDAC is often challenging because cancer symptoms will progress only at later stages (advanced stage). One of the earlier symptoms of PDAC is Jaundice. Patients with diabetes, obesity, and alcohol consumption are also at higher risk of having pancreatic cancer. A decision support system is developed to detect pancreatic cancer at an earlier stage to address this challenge. Features such as Mean Hue, Mean Saturation, Mean Value, and Mean Standard Deviation are computed after color space conversion from RGB to HSV. Fuzzy k-Nearest Neighbor (F-kNN) is designed for classification. The system proposed is trained and tested using features extracted from jaundiced eye images. The proposed system results indicate that this model can predict pancreatic cancer as earlier as possible, helping clinicians make better decisions for surgical planning.

show abstract

“…Po standardni metodologiji ocenjevanja uporabljamo štiri merila uspešnosti za poročanje o uspešnosti za segmentacijske naloge, to so Jaccardov indeks IoU, natančnost, priklic in mera F1 (Rot, 2020;Emeršič, 2021).…”

Section: Metrike Uspešnostiunclassified

ROSUS 2022 - Računalniška obdelava slik in njena uporaba v Sloveniji 2022: Zbornik 16. strokovne konference

2022

View full text Add to dashboard Cite

Povzetek V zadnjem desetletju se je v Laboratoriju za računalniški vid na FRI UL oblikovala močna skupina, ki dela na področju biometrije. Skupina je močno povezana z Laboratorijem za strojno inteligenco na FE UL. Prvi ključni koraki so bili narejeni v okviru kompetenčnih centrov, kjer smo v oblaku naredili fuzijo modalnosti, obrazov in prstih odtisov. Vzporedno s tem se je odvijalo delo na razpoznavanju ljudi iz načina gibanja. Nato je delo na področju biometrije dobilo še dodaten zagon. Posvetili smo se povsem novi modalnosti uhljev, začeli delati na izzivu fotorealistične deidentifikacije obrazov, dodali beločnico, šarenico ter obočesno regijo kot naslednje tri sveže modalnosti. V zadnjem času se ukvarjamo tudi z detekcijo globokih ponaredkov, oceno pravilnega nošenja mask, virtualnim pomerjanjem oblačil, obdelavo prstnih sledi (latentov) in celovitim ohranjanjem zasebnosti obrazov. Na drugi stopnji študija smo uvedli tudi nov izbirni predmet Slikovna biometrija. Ta ima letos spet skoraj 90 slušateljev. Število članov skupine trenutno raste iz leta v leto, temu primerno tudi financiranje in publikacije na ključnih konferencah ter v revijah, nenazadnje pa se vpliv skupine pozna tudi pri organizaciji tekmovanj na teh konferencah ter tudi zmagah na sorodnih tekmovanjih. Predavanje bo osvetlilo prehojeno pot skozi ključne raziskovalne vsebine. Spletna stran: https://fri.uni-lj.si/sl/laboratorij/lrv.

show abstract

Deep Sclera Segmentation and Recognition

Cited by 44 publications

References 73 publications

Learning to Combine Local and Global Image Information for Contactless Palmprint Recognition

Learning to Combine Local and Global Image Information for Contactless Palmprint Recognition

Early Detection of Pancreatic Cancer Using Jaundiced Eye Images

ROSUS 2022 - Računalniška obdelava slik in njena uporaba v Sloveniji 2022: Zbornik 16. strokovne konference

Contact Info

Product

Resources

About