Fast selection of grayvalues within a given gray intervall.
fast_threshold selects the pixels from the input image whose gray values g fulfill the following condition:
MinGray <= g <= MaxGray .
To reduce procesing time, the selection is done in two steps: At first all pixels along rows with distances MinHeight are processed. In the next step the neighborhood (size MinHeight x MinHeight) of all previously selected points are processed.
|
Image (input_object) |
image(-array) -> object : byte / uint2 / direction / cyclic |
| Image to be thresholded. | |
|
Region (output_object) |
region(-array) -> object |
| Regions with gray values lying in the specified interval. | |
|
MinGray (input_control) |
number -> real / integer |
| Lower threshold for the gray values. | |
| Default value: 128 | |
| Suggested values: 0.0, 10.0, 30.0, 64.0, 128.0, 200.0, 220.0, 255.0 | |
| Typical range of values: 0.0 <= MinGray <= 255.0 (lin) | |
| Minimum increment: 1 | |
|
Recommended increment: 5.0 | |
|
MaxGray (input_control) |
number -> real / integer |
| Upper threshold for the gray values. | |
| Default value: 255.0 | |
| Suggested values: 0.0, 10.0, 30.0, 64.0, 128.0, 200.0, 220.0, 255.0 | |
| Typical range of values: 0.0 <= MaxGray <= 255.0 (lin) | |
| Minimum increment: 1 | |
|
Recommended increment: 5.0 | |
|
MinHeight (input_control) |
number -> integer |
| Minimum height of objects to be extracted. | |
| Default value: 20 | |
| Suggested values: 5, 10, 15, 20, 25, 30, 40, 50, 60, 70, 100 | |
| Typical range of values: 2 <= MinHeight <= 200 (lin) | |
| Minimum increment: 1 | |
|
Recommended increment: 2 | |
Let A be the area of the ouput region and height the height of Image. Then the runtime complexity is O(A + height / MinHeight).
fast_threshold returns 2 (H_MSG_TRUE) if all parameters are correct. The behavior with respect to the input images and output regions can be determined by setting the values of the flags 'no_object_result', 'empty_region_result', and 'store_empty_region' with set_system. If necessary, an exception is raised.
fast_threshold is reentrant and automatically parallelized (on tuple level).
histo_to_thresh, min_max_gray, sobel_amp, gauss_image, reduce_domain, fill_interlace
connection, dilation1, erosion1, opening, closing, rank_region, shape_trans, skeleton
class_2dim_sup, hysteresis_threshold, dyn_threshold
Region processing
