Image Connected Component Analysis
Function: Performs connected component analysis on a binarized image to obtain a label map for each connected component, thereby deriving detection mask results.
Input Parameters:
| Name | Type | Valid Range | Default Value | Meaning |
|---|---|---|---|---|
Binarized image |
BinaryImage |
None |
None |
Binarized image |
Output Parameters:
| Name | Type | Valid Range | Default Value | Meaning |
|---|---|---|---|---|
Detection results |
DetectInstance |
None |
{} |
Returns bounding boxes, categories, scores and polygon. |
Results center point list |
List |
None |
None |
List of result center points, list of n * 2. |
Result mask image |
BinaryImage |
None |
None |
Generate multiple mask result maps for each input image. |
Parameter Settings:
| Name | Type | Valid Range | Default Value | Meaning |
|---|---|---|---|---|
Result type |
String |
['polygon', 'minAreaRect'] |
minAreaRect |
In the returned detection results, the type of polygon field, Polygon is polygon, MinAreaRect is the minimum bounding box. |
Connected domain type |
Integer |
[4, 8] |
8 |
connected domains, 8 connected domains or 4 connected domains. |
Minimum pixel area |
Integer |
[5, 400000] |
100 |
The smallest pixel area, which filters out results with pixel areas smaller than this value. |
Maximum pixel area |
Integer |
[5, 400000] |
300000 |
The maximum pixel area, which filters out results with pixel area greater than this value. |
Return mask list |
Bool |
None |
False |
Whether to return a list of masks. |