Article Figures & Data
Figures
- FIGURE 1.
Noiseless computer simulation with 2 hot lesions. First row shows images, and second row shows central horizontal profiles of corresponding images in first row. Left: input blurred noiseless image that contains 2 hot pointlike lesions and positive background. Middle: result by direct application of regular Lucy–Richardson algorithm. Right: result of proposed segmentation-filtering method.
- FIGURE 2.
Noisy computer simulation with 2 hot lesions. First row shows images, and second row shows central horizontal profiles of corresponding images in first row. Left: input blurred noisy image that contains 2 hot pointlike lesions and positive background. Middle: result by direct application of regular Lucy–Richardson algorithm. Right: result of proposed segmentation-filtering method.
- FIGURE 3.
Noiseless computer simulation with 2 cold lesions. First row shows images, and second row shows central horizontal profiles of corresponding images in first row. Left: input blurred noiseless image that contains 2 cold pointlike lesions and positive background. Middle: result by direct application of regular Lucy–Richardson algorithm. Right: result of proposed segmentation-filtering method.
- FIGURE 4.
Noisy computer simulation with 2 cold lesions. First row shows images, and second row shows central horizontal profiles of corresponding images in first row. Left: input blurred noisy image that contains 2 cold pointlike lesions and positive background. Middle: result by direct application of regular Lucy–Richardson algorithm. Right: result of proposed segmentation-filtering method.
- FIGURE 5.
PET 18F NEMA phantom experiment. First column shows images, second column shows profile along upper horizontal line (passing through cold lesion, cold central cylinder, and hot lesion), and third column shows profile along lower horizontal line (passing through 2 hot lesions). Top row: input blurred noisy raw reconstruction image. Middle row: result of proposed segmentation-filtering method. Bottom row: result by direct application of regular Lucy–Richardson algorithm.
