Review Article

Ear Biometrics Using Deep Learning: A Survey

Table 8

Cont. differences between this review article and the recent/existing review papers.

Author(s) and date of publicationPaper titleAim/focus/objectivePaper coverage (year) and scope

(5) Zhang and Mu [65] 24 January 2017Ear detection under uncontrolled conditions with multiple scale faster region-based convolutional neural networksThis system contained large occlusions, scale, and pose variationForty-one (41) application papers that are deep learning ear identification methods are reviewed in this paper

(6) Kohlakala and Coetzer [66] 1 June 2021Ear-based biometric authentication through the detection of prominent contourIt is used to classify ears either in the foreground or background of the image. The binary contour image applied the matching for feature extraction, and this was carried out by implementing Euclidean distance measure, which had a ranking to verify for authenticationTwenty-one (21) application papers that are deep learning ear identification methods are reviewed in this paper

(7) Tomczyk and Szczepaniak [67] 13 December 2019Ear detection using convolutional neural network on graphs with filter rotationIt shows the published experimental results that the approach performed the rotation equivalence property to detect rotated structuresForty (40) application papers that are deep learning ear identification methods are reviewed in this paper

(8) Alshazly et al. [68] 8 December 2019Handcrafted versus CNN features for ear recognitionThe paper took seven performing handcrafted descriptors to extract the discriminating ear image. They then took the extracted ear and trained it using Support Vector Machines (SVM) to learn a suitable modelSeventy-three (73) application papers that are deep learning ear identification methods are reviewed in this paper