Hybrid Segmentation Scheme for Skin Features Extraction Using Dermoscopy Images
Source abstract
Abstract
Objective and quantitative assessment of skin conditions is essential for cosmeceutical studies and research on skin aging and skin regeneration. Various handcraft-based image processing methods have been proposed to evaluate skin conditions objectively, but they have unavoidable disadvantages when used to analyze skin features accurately. This study proposes a hybrid segmentation scheme consisting of Deeplab v3+ with an Inception-ResNet-v2 backbone, LightGBM, and morphological processing (MP) to overcome the shortcomings of handcraft-based approaches. First, we apply Deeplab v3+ with an Inception-ResNet-v2 backbone for pixel segmentation of skin wrinkles and cells. Then, LightGBM and MP are used to enhance the pixel segmentation quality. Finally, we determine several skin features based on the results of wrinkle and cell segmentation. Our proposed segmentation scheme achieved a mean accuracy of 0.854, mean of intersection over union of 0.749, and mean boundary F1 score of 0.852, which achieved 1.1%, 6.7%, and 14.8% improvement over the panoptic-based semantic segmentation method, respectively.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Cutaneous Melanoma Detection and Management · Industrial Vision Systems and Defect Detection · Textile materials and evaluations
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (44)
Show source-provided references
1. Skin ageing10.1111/j.1468-3083.2010.03963.x ↗
2. Intrinsic and extrinsic factors in skin ageing: A review10.1111/j.1468-2494.2007.00415.x ↗
3. Introduction to skin aging10.1016/j.jtv.2016.03.002 ↗
4. Beyond a glimpse at seasonal dry skin: A review10.1159/000047984 ↗
5. Age, sunlight, and facial skin: A histologic and quantitative study10.1016/S0190-9622(08)80964-4 ↗
6. Skin topography measurement by interference fringe projection: A technical validation10.1034/j.1600-0846.2001.70210.x ↗
7. Age-related changes in skin wrinkles assessed by a novel three-dimensional morphometric analysis10.1046/j.1365-2133.2002.04874.x ↗
8. A study of differences in surface roughness between sun-exposed and unexposed skin with age10.1034/j.1600-0781.2003.00042.x ↗
9. Topography and anisotropy of the skin surface with ageing10.1111/j.1600-0846.2005.00096.x ↗
10. Skin texture modeling10.1007/s11263-005-4637-2 ↗
11. Human skin texture analysis using image processing techniques
12. Skin texture aging trend analysis using dermoscopy images10.1111/srt.12143 ↗
13. Age-dependent skin texture analysis and evaluation using mobile camera image10.1111/srt.12459 ↗
14. Skin aging estimation scheme based on lifestyle and dermoscopy image analysis10.3390/app9061228 ↗
15. Global and systematic demonstration for the practical usage of a direct in vivo measurement system to evaluate wrinkles10.1111/j.1468-2494.2007.00399.x ↗
16. An analysis of human dorsal hand skin texture using hyperspectral imaging technique for assessing the skin aging process10.1177/0003702816659667 ↗
17. Age-dependent changes in skin surface assessed by a novel two-dimensional image analysis10.1111/j.1600-0846.2009.00377.x ↗
18. Encoder-decoder with atrous separable convolution for semantic image segmentation
19. Inception-v4, inception-ResNet and the impact of residual connections on learning
20. Lightgbm: A highly efficient gradient boosting decision tree
21. Three-dimensional morphological characterization of the skin surface micro-topography using a skin replica and changes with age10.1111/srt.12119 ↗
22. 3D in-vivo optical skin imaging for topographical quantitative assessment of non-ablative laser technology
23. Accuracy and precision of silicon based impression media for quantitative areal texture analysis10.1038/srep10800 ↗
24. Assessing facial wrinkles: Automatic detection and quantification10.1111/j.1600-0846.2012.00635.x ↗
25. Quantitative evaluation of elderly skin based on digital image analysis10.1111/j.1600-0846.2007.00278.x ↗
26. Age range estimation based on facial wrinkle analysis using hessian based filter
27. Skin feature extraction and processing model for statistical skin age estimation10.1007/s11042-011-0987-7 ↗
28. Monitoring skin condition using life activities on the SNS user documents10.1007/s11042-017-4623-z ↗
29. Skin lesion classification: An optimized framework of optimal color features selection
30. A hierarchical three-step superpixels and deep learning framework for skin lesion classification
31. Pixels to classes: Intelligent learning framework for multiclass skin lesion localization and classification10.1016/j.compeleceng.2020.106956 ↗
32. Attributes based skin lesion detection and recognition: A mask RCNN and transfer learning-based deep learning framework10.1016/j.patrec.2020.12.015 ↗
33. A novel framework for rapid diagnosis of COVID-19 on computed tomography scans
34. Handcrafted vs. non-handcrafted features for computer vision classification10.1016/j.patcog.2017.05.025 ↗
35. The watershed transform: Definitions, algorithms and parallelization strategies10.3233/FI-2000-411207 ↗
36. Group-sparse representation with dictionary learning for medical image denoising and fusion10.1109/TBME.2012.2217493 ↗
37. Automatic liver segmentation for volume measurement in CT Images10.1016/j.jvcir.2005.07.001 ↗
38. Generative adversarial network in medical imaging: A review10.1016/j.media.2019.101552 ↗
39. Wrinkle feature-based skin age estimation scheme
40. Performance comparison of median and wiener filter in image de-noising10.5120/1664-2241 ↗
41. What is a good evaluation measure for semantic segmentation?
42. U-net: Convolutional networks for biomedical image segmentation
43. SegNet: A deep convolutional encoder-decoder architecture for image segmentation10.1109/TPAMI.2016.2644615 ↗
44. Panoptic feature pyramid networks