Tongue Segmentation Using Active Contour Model

Saparudin, Saparudin; Erwin, Erwin; Fachrurrozi, Muhammad

doi:10.11591/eecsi.v3i1.1116

Cited by 2 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To extend the clustering methods onto various color spaces, Wang et al 19 propose a two‐stage clustering method termed fuzzy Kohonen clustering networks, which is combined with genetic algorithms to optimize the interclass and the intraclass distances of the clusters to generate segmented tongue maps. In the alternate view, the active contour model 20 or Snake 5,21,22 is one of the most representative methods, which iteratively fits an elastic shrinkage by optimizing an energy function. In Reference 5, a rough initial contour with the key points of the tongue body are first obtained by prior knowledge, and then Snake is implemented to direct curve evolution of the tongue contour with color and gradient flow field for the upper part and the lower part of the tongue, respectively.…”

Section: Related Workmentioning

confidence: 99%

A robust interclass and intraclass loss function for deep learning based tongue segmentation

Cai

Wang

Liu

et al. 2020

Concurrency and Computation

View full text Add to dashboard Cite

The fast and robust segmentation of tongue images is a prerequisite to achieve automatic tongue diagnosis in traditional Chinese medicine. In order to assist tongue diagnosis in real-life scenarios, an ideal tongue segmentation method would need to obtain the entire tongue body as well as its precise contours. However, the similar appearance among the tongue body, the coating, and the lips hinders the performance for most unsupervised learning methods that primarily utilize low-level visual features. On the other hand, although the supervised deep convolutional neural networks (DCNNs) that typically depend on the widely used crossentropy loss can achieve better accuracy, they are very prone to segment image to multiple trivially areas. To address both of the above issues, we make an attempt to boost the segmentation performance of DCNNs with a novel auxiliary loss function that seeks to exploit large margin learning for end-to-end tongue segmentation models. Specifically, we first propose a loss function that involves interclass and intraclass costs to directly measure the distance among pixels that belong to different connected regions. Then, we explore the potential ability of this loss function as a regularization for different segmentation networks such as those with attention modules and the deeply supervised network. Finally, a theoretical analysis for this learning scheme is presented. Through experiments on challenging datasets, we show that the proposed approach can be easily integrated into state-of-the-art networks to boost their performance at the tongue segmentation task without bells and whistles.

show abstract

Section: Related Workmentioning

confidence: 99%

A robust interclass and intraclass loss function for deep learning based tongue segmentation

Cai

Wang

Liu

et al. 2020

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Individuals and identical twins have noteworthy variances in the tongue's defining characteristics. It is no surprise that the tongue has its network of nerves, muscles, and blood arteries, just like every other organ [1]. It also has taste buds and papillae.…”

Section: Introductionmentioning

confidence: 99%

“…The tongue is a fairly reliable form of human recognition identification and can act as a generic biometric in any given application. The tongue is an essential organ in the human body and has many uses, including moving food for chewing and tasting and helping make speech sounds [1]. A digital device (camera) is used to collect tongue images.…”

Section: Introductionmentioning

confidence: 99%

People identification via tongue print using fine-tuning deep learning

Obaid,

Y. Kamil,

Hamza

2023

IJRES

View full text Add to dashboard Cite

<span>Many person-verification systems are critical in security systems for verifying passage through doors opened to specific people using various techniques. People can use electronic payment methods and security apps to generate codes for quick, remote financial transactions. Older systems required precision and speed. Many alternative methods were developed by technology and artificial intelligence to make such operations simple and quick. The identification of tongue prints is discussed in this paper. Tongue prints, like fingerprints, are unique to each individual. The tongue was used in this study because it is unique among such organs. The tongue is protected by the lips. This guards against taking a tongue print by force. Some people distort their fingerprints, making fingerprint recognition systems unable to recognize them. Car accidents cause facial distortion, which distorts the system and prevents it from distinguishing facial prints, so the tongue was used as a fingerprint in this study. A database of 1,104 images for 138 Mustansiriyah University College of Science students yielded an average of eight images per individual. VGG16 was implemented for transfer learning and fine-tuning. In comparison to previous studies, the accuracy achieved was more than 91%.</span>

show abstract

Tongue Segmentation Using Active Contour Model

Abstract: Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

Cited by 2 publications

References 0 publications

A robust interclass and intraclass loss function for deep learning based tongue segmentation

A robust interclass and intraclass loss function for deep learning based tongue segmentation

People identification via tongue print using fine-tuning deep learning

Contact Info

Product

Resources

About