dyn_threshold

dyn_threshold ( OrigImage, ThresholdImage : RegionDynThresh : Offset, LightDark : )

Segment an image using a local threshold.

dyn_threshold selects from the input image those regions in which the pixels fulfill a threshold condition. Let g_{o} = g_{OrigImage}, and g_{m} = g_{ThresholdImage}. Then the condition for LightDark = 'light' is:

                 g_o >= g_m + Offset

For LightDark = 'dark' the condition is:

                 g_o <= g_m - Offset

For LightDark = 'equal' it is:

                 g_m - Offset <= g_o <= g_m + Offset

Finally, for LightDark = 'not_equal' it is:

                 g_m - Offset > g_o oder g_o > g_m + Offset

This means that all points in OrigImage whose gray value is larger then or equal to the gray value in ThresholdImage plus an offset are aggregated into the resulting region.

Typically, the threshold images are smoothed versions of the original image (e.g., by applying mean_image, gauss_image, etc.). Then the effect of dyn_threshold is similar to applying threshold to a highpass-filtered version of the original image (see highpass_image).

With dyn_threshold contours of an object can be extracted, where the objects' size (diameter) is determined by the mask size of the lowpass filter and the amplitude of the objects' edges:

The larger the mask size is chosen, the larger the found regions get. As a rule of thumb, the mask size should be about twice the diameter of the objects to be extracted. It is important not to set the parameter Offset to zero because in this case too many small regions will be found (noise). Values between 5 and 40 are a sensible choice. The larger Offset is chosen, the smaller the extracted regions get.

All points of the input image fulfilling the above condition are stored jointly in one region. If necessary, the connected components can be obtained by calling connection.

Attention

If Offset is chosen from -1 to 1 usually a very noisy region is generated, requiring large storage. If Offset is chosen too large (> 60, say) it may happen that no points fulfill the threshold condition (i.e., an empty region is returned). If Offset is chosen too small (< -60, say) it may happen that all points fulfill the threshold condition (i.e., a full region is returned).

Parameters

`OrigImage` (input_object)	image(-array) `->` object : byte / int2 / uint2 / int4 / real
Image to be segmented.

`ThresholdImage` (input_object)	image(-array) `->` object : byte / int2 / uint2 / int4 / real
Image containing the local thresholds.

`RegionDynThresh` (output_object)	region(-array) `->` object
Segmented regions.

`Offset` (input_control)	number `->` real / integer
Offset added to ThresholdImage.
Default value: 5.0
Suggested values: 1.0, 3.0, 5.0, 7.0, 10.0, 20.0, 30.0
Typical range of values: `-255.0 <= Offset <= 255.0` (lin)
Minimum increment: 0.01
Recommended increment: 5
Restriction: `(-255 < Offset) && (Offset < 255)`

`LightDark` (input_control)	string `->` string
Extract light, dark or similar areas?
Default value: 'light'
List of values: 'dark', 'light', 'equal', 'not_equal'

Example

/* Looking for regions with the diameter D */
mean_image(Image,Mean,D*2+1,D*2+1) 
dyn_threshold(Image,Mean,Seg:5,'light') 
connection(Seg,Regions).

Complexity

Let F be the area of the input region. Then the runtime complexity is O(F).

Result

dyn_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.

Parallelization Information

dyn_threshold is reentrant and automatically parallelized (on tuple level, domain level).

Possible Predecessors

mean_image, smooth_image, gauss_image

Possible Successors

connection, select_shape, reduce_domain, select_gray, rank_region, dilation1, opening

Alternatives

check_difference, highpass_image, sub_image, threshold

Module

Region processing