Tongue Color Classification in TCM with Noisy Labels via Confident-Learning-Assisted Knowledge Distillation

Abstract

Tongue color is an important tongue diagnostic index for traditional Chinese medicine (TCM). Due to the individual experience of TCM experts as well as ambiguous boundaries among the tongue color categories, there often exist noisy labels in annotated samples. Deep neural networks trained with the noisy labeled samples often have poor generalization capability because they easily overfit on noisy labels. A novel framework named confident-learning-assisted knowledge distillation (CLA-KD) is proposed for tongue color classification with noisy labels. In this framework, the teacher network plays two important roles. On the one hand, it performs confident learning to identify, cleanse and correct noisy labels. On the other hand, it learns the knowledge from the clean labels, which will then be transferred to the student network to guide its training. Moreover, we elaborately design a teacher network in an ensemble manner, named E-CA 2 -ResNet18, to solve the unreliability and instability problem resulted from the insufficient data samples. E-CA 2 -ResNet18 adopts ResNet18 as the backbone, and integrates channel attention (CA) mechanism and activate or not activation function together, which facilitates to yield a better performance. The experimental results on three self-established TCM tongue datasets demonstrate that, our proposed CLA-KD can obtain a superior classification accuracy and good robustness with a lower network model complexity, reaching 94.49%, 92.21%, 93.43% on the three tongue image datasets, respectively.

Stereo matching from monocular images using feature consistency

“…Before synthesizing the corresponding stereo pairs from new viewpoints using forward warping, we take care to solve the problems associated with occlusions and artefacts. Recent studies have shown that pre-training a significant amount of data with noisy labels can enhance the performance of image classification [21][22][23][24]. Inspired by this, we applied a similar approach to train a stereo matching network.…”