Deep-segmentation of plantar pressure images incorporating fully convolutional neural networks

Wang, Dan; Li, Zairan; Dey, Nilanjan; Ashour, Amira S.; Moraru, Luminița; Sherratt, R. Simon; Shi, Fuqian

doi:10.1016/j.bbe.2020.01.004

Cited by 20 publications

(7 citation statements)

References 26 publications

(24 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, import the original measurement data of the female last (the data of the female last comes from empirical data); use the footprint method to form a relatively matching foot corresponding model. Then, the last is repaired and refined by manual experience, and a one-size shoe last is produced, which is usually completed by multiple shoe last designers [15][16]. After the mother last is completed, it is scanned in 3D and imported into the last body design software for meshing operation to provide basic data models for last body trimming, different shoe size expansion and contraction, paired custom processing, and mass production processing.…”

Section: Resultsmentioning

confidence: 99%

Design Elements Extraction from Point-Cloud Dataset Using Deep Enhancement Learning

Wang

2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Recent research has focused on learning 3D data through deep learning and improving geometric processing methods. Two aspects distinguish deep learning of point clouds from deep learning of images. A core aspect of the point cloud data is that it is disordered in comparison to the 2D image, so when designing the network, we need to consider how to deal with it. Reinforcement learning maximizes the expectation of cumulative rewards. The point cloud data was used for modeling and applied to the sole surface construction to achieve better results.

show abstract

Section: Resultsmentioning

confidence: 99%

Design Elements Extraction from Point-Cloud Dataset Using Deep Enhancement Learning

Wang

2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

show abstract

“…The talar leg joint is painful and has poor stability. There may be knee hyperextension in the early support phase, lack of strength in pedaling, and limb clearance obstacles in the swing phase [31]. In this deformity, the affected foot rotates inward at the ankle-the foot points down, facing inward and the sole.…”

Section: Figure1 the Experiments On Plantar Pressure Dataset Acquisimentioning

confidence: 99%

An Improved Residual-Network Model-based Conditional Generative Adversarial Network Plantar Pressure Image Classification: A Comparison of Normal, Planus, and Talipes Equinovarus Feet

Han

Wang

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The number of layers of deep learning (DL) increases, and following the performance of computing nodes improvement, the output accuracy of deep neural networks (DNN) faces a bottleneck problem. The resident network (RN) based DNN model was applied to address these issues recently. This paper improved the RN and developed a rectified linear unit (ReLU) based conditional generative adversarial nets (cGAN) to classify plantar pressure images. A foot scan system collected the plantar pressure images, in which normal (N), planus (PL), and talipes equinovarus feet (TE) data-sets were acquired subsequently. The 9-foot types named N, PL, TE, N-PL, N-TE, PL-N, PL-TE, TE-N, and TE-PL were classified using the proposed DNN models, named resident network-based conditional generative adversarial nets (RNcGAN). It improved the RN structure firstly and the cGAN system hereafter. In the classification of plantar pressure images, the pixel-level state matrix can be direct as an input, which is different from the previous image classification task with image reduction and feature extraction. cGAN can directly output the pixels of the image without any simplification. Finally, the model achieved better results in the evaluation indicators of accuracy (AC), sensitivity (SE), and F1-measurement (F1) by comparing to artificial neural networks (ANN), k-nearest neighbor (kNN), Fast Region-based Convolution Neural Network (Fast R-CNN), visual geometry group (VGG16), scaled-conjugate-gradient convolution neural networks (SCG-CNN), GoogleNet, AlexNet, ResNet-50-177, and Inception-v3. The final prediction of class accuracy is 95.17%. Foot type classification is vital for producing comfortable shoes in the industry.

show abstract

“…The convolutional neural network based recognition methods demonstrate unique advantages and powerful capabilities in recent years [9]- [11]. Wang et al [12] introduced a fully convolutional networks based plantar pressure image segmentation method, the SegNet is utilized as the backbone. In [12], the researcher also found that the plantar pressure image segmentation results have potential applications in improving the comfort of shoes.…”

Section: Introductionmentioning

confidence: 99%

“…Wang et al [12] introduced a fully convolutional networks based plantar pressure image segmentation method, the SegNet is utilized as the backbone. In [12], the researcher also found that the plantar pressure image segmentation results have potential applications in improving the comfort of shoes. An AlexNet based plantar pressure image segmentation method is also proposed in [13], and the computation complexity of plantar pressure sensor datasets is decreased via the segmentation method.…”

Section: Introductionmentioning

confidence: 99%

Diabetes Noninvasive Recognition via Improved Capsule Network

WANG

2022

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Noninvasive recognition is an important trend in diabetes recognition. Unfortunately, the accuracy obtained from the conventional noninvasive recognition methods is low. This paper proposes a novel Diabetes Noninvasive Recognition method via the plantar pressure image and improved Capsule Network (DNR-CapsNet). The input of the proposed method is a plantar pressure image, and the output is the recognition result: healthy or possibly diabetes. The ResNet18 is used as the backbone of the convolutional layers to convert pixel intensities to local features in the proposed DNR-CapsNet. Then, the PrimaryCaps layer, SecondaryCaps layer, and DiabetesCaps layer are developed to achieve the diabetes recognition. The semantic fusion and locality-constrained dynamic routing are also developed to further improve the recognition accuracy in our method. The experimental results indicate that the proposed method has a better performance on diabetes noninvasive recognition than the state-of-the-art methods.

show abstract

Deep-segmentation of plantar pressure images incorporating fully convolutional neural networks

Cited by 20 publications

References 26 publications

Design Elements Extraction from Point-Cloud Dataset Using Deep Enhancement Learning

Design Elements Extraction from Point-Cloud Dataset Using Deep Enhancement Learning

An Improved Residual-Network Model-based Conditional Generative Adversarial Network Plantar Pressure Image Classification: A Comparison of Normal, Planus, and Talipes Equinovarus Feet

Diabetes Noninvasive Recognition via Improved Capsule Network

Contact Info

Product

Resources

About