• Home
  • Lab Facilities
  • User Guide
  • Remote Experiment
  • FAQ
  • vlab
  • Feedback

Computer Integrated Manufacturing System Laboratory

Funded by Ministry of Human Resource and Development, Govt. of India

Blob Analysis by Segmentation and Thresolding :

It is one of the most important features for automated vision system, since it is a stage of processing that objects are extracted from a scene for subsequent recognition and analysis.

  • Snap an image by view flex camera. Click on BLOB analysis
  • Following window will open.
  • Go to segmentation tab

 

Segmentation: Segmentation process subdivides a sensed image into constituent parts or objects. It is one of the most important elements of automated vision system, since it is at this stage of processing that objects are extracted from a scene for subsequent recognition and analysis, which is most important process to analysis the blob of an image.

Segmentation Tab Options

  • Count:  Performs the same operation as Analysis Blob Count on the created mask.
  • Foreground: Specifies whether black pixels or white pixels are considered foreground (or blob) pixels.
  • Source gray: Displays the name of the source image and ROI.
  • Identification Mask: Allows you to select a segmentation mask.

Threshold current image: Uses a threshold version the source image/ROI as the mask. This is the usual mode.
Same as current image: Uses the current image/ROI as the mask. For example, in labeled identification mode, you would not want a thresholded mask.
Copy the selected image: Makes an internal copy of the currently active image/ROI as the mask. When using this option, the blob identifier mask and gray source image are completely independent of each other.
Threshold: In its simplest form, thresholding is a binary conversion technique in which each pixel is converted into a binary value, either black or white. Thresholding can be used to binarize an image. Binarizing reduces an image to two grayscale values (for example, 0 and
255, as below). Note that binary images can be used as a mask to identify blobs in a blob analysis application
Specifies the threshold values when using a Threshold current image mask, otherwise disabled. Enter the low and high threshold values for the operation. You can either type in these values or move the slider bars. The pixel intensities between the two sliders are set to black (pixel values are set to 0) and the others are set to white (pixel values are set to 255). As you change the threshold values, the blob identifier image is updated.

Auto: Sets the threshold values automatically.
Identification Mode: Specifies how the blobs are to be measured.

    • Individual: Measures all blobs individually.
    • Whole image: Measures all blobs as a group. Blobs in the image are treated as one blob and features are calculated for this grouped blob.
    • Labeled: Measures blobs with the same label value as a group. When using Labeled mode, ensure that each blob in the identifier image has a uniform pixel value.
             Previous           Index          Next