Start » Filter Reference » OpenCV » Miscellaneous Image Transformations » cvAdaptiveThreshold
Basic| Module: | OpenCV |
|---|
Applies an adaptive threshold to an image.
| Name | Type | Range | Description | |
|---|---|---|---|---|
 |
inSrc | Image | UINT8, single-channel input image. | |
|
inMaxValue | Real | Non-zero value assigned to the pixels for which the condition is satisfied. | |
|
inAdaptiveMethod | CvAdaptiveMethod | Adaptive thresholding algorithm to use. | |
|
inThresholdType | CvThresholdType | Thresholding type. | |
|
inBlockSize | Integer | 3 -  |
Size of a pixel neighborhood that is used to calculate a threshold value for the pixel: 3, 5, 7, and so on. |
|
inC | Real | Constant subtracted from the mean or weighted mean | |
 |
outDst | Image | Output image. | |
Description
The operation transforms each pixel value to the maximum or minimum level thus creating binary image. The result of the transformation depends on the relative pixel intensity:
- Pixel values that are brighter than local average of the pixel neighbourhood are transformed to inMaxValue.
- Other pixel values are transformed to 0.
Pixel neighbourhood used to compute the local average is a rectangle of dimensions (inBlockSize) x (inBlockSize) centered at the pixel being processed.
Local average is computed according to method specified in inAdaptiveMethod:
- For the method ADAPTIVE_THRESH_MEAN, the threshold value is mean of pixels in neighbourhood minus inC.
- For the method ADAPTIVE_THRESH_GAUSSIAN, the threshold value is a weighted sum of values in pixels in neighbourhood minus inC. The weights are computed using normal distribution with default sigma value for specified inBlockSize
Examples
 |
 |
cvAdaptiveThreshold performed on the sample image with inAdaptiveMethod = ADAPTIVE_THRESH_GAUSSIAN, inThresholdType = THRESH_BINARY, inBlockSize = 13, inC = -5.0.
Errors
This filter can throw an exception to report error. Read how to deal with errors in Error Handling.
List of possible exceptions:
| Error type | Description |
|---|---|
| DomainError | Input image must be UINT8 single-channel in cvAdaptiveThreshold. |
Complexity Level
This filter is available on Basic Complexity Level.