Fig. 1. Overview of the systematically analyzing method of lipid images

Fig. 2. The procedure of pre-processing

Fig. 3. Lipid region segmentation results based on the LGIF model. (a) Intensity value; (b) segmentation contours (red rectangles mark incorrect segmentation areas); (c) segmentation results

Fig. 4. (a) The pixel’s hue value; (b) the pixel’s intensity value; (c) the modified image

Fig. 5. Experimental procedure

Fig. 6. The prototype of the endocytoscopic imaging system. (a) The mobile workstation, including the lightbox, the video system center, and endocytoscope; (b) the knob for amplification and attitude change; (c) structure of light source; (d) the tip of the endocytoscope

Fig. 7. The experimental subjects. (a) The phantom obtained by demoulding from the 3D-printing model; (b) the phantom covered with lipid; (c) a comparison of the reflection spectra between the phantom and pig stomach

Fig. 8. The color, hue value, and intensity images of (a) NBI and (b)WLI

Fig. 9. (a) The enhanced images, (b) reflective detection results, and (c) inpainting results

Fig. 10. Enhancement and segmentation results. (a) Initial images; (b) segmentation results; (c) manual annotations

Fig. 11. Segmentation results of triangular initial contour. (a) Initial contour; (b) incorrect segmentation results (blue lines mark the segmentation boundary)

Fig. 12. Segmentation results using different weight factors

Fig. 13. (a) Input images; segmentation results obtained by (b) C-V model; (c) LBF model; (d) LGIF model; (e) the proposed method; (f) manual annotations

Fig. 14. Segmentation of WOS images in previous papers^{[10, 12]}. (a) Initial images; (b) segmentation contours; (c) segmentation results