Image region partitioning using pre-labeled regions

• US 7,580,556 B2
• Filed: 01/26/2004
• Issued: 08/25/2009
• Est. Priority Date: 01/26/2004
• Status: Active Grant

• Current Assignee: Leica Microsystems Gmbh (Danaher Corporation)
• Original Assignee: DRVision Technologies, LLC (Danaher Corporation)
• Inventors: Lee, Shih-Jong J., Phan, Tuan
• Primary Examiner(s): Strege; John B
• Application Number:
  US10/767,530
• Publication Number:
  US 20050163373A1
• Time in Patent Office: 2,038 Days
• Field of Search: 382/173, 382/133
• US Class Current: 382/133
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/20041 Distance transform
• CPC Class Codes: G06T 2207/20044 Skeletonization; Medial axis transform
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/155 involving morphological operators
• CPC Class Codes: G06T 7/187 involving region growing; involving region merging; involving connected component labelling
• CPC Class Codes: G06T 7/60 Analysis of geometric attributes
• CPC Class Codes: G06V 10/457 by analysing connectivity, e.g. edge linking, connected component analysis or slices
• CPC Class Codes: G06V 20/695 Preprocessing, e.g. image segmentation

Abstract

A two pass Zone Of Influence (ZOI) creation method creates a Zone Of Influence (ZOI) image. The two pass ZOI creation method performs a first pass scan to create a first pass intermediate distance image and a shortest distance component label image. It then performs a second pass scan using the first pass intermediate distance image and the shortest distance component label image to create a background distance transform image and a updated shortest distance component label image. An adaptive image region partition method allows distance lengths of the two pass adaptive ZOI creation method depend on their associated component labels.

Claims

1. An adaptive image region partition method using a computer or processor to carry out the steps of:

a) Input a component labeled image and its associated characteristics for each component;

b) Perform an adaptive two pass ZOI creation method using the component labeled image and component characteristics to create an adaptive ZOI image using a length function I wherein a component specific weighting factor w is applied to the length function to generated a component weighted length;

l[(i,j),Z(i,j)]=l(i,j)*w(Z(i,j))

wherein w is a general function of component characteristics that is related to one of component size, component confidence, or types of components and Z(i,j) is component labeled image value at image position (i,j).

2. The adaptive image region partition method of claim 1 wherein the adaptive two pass ZOI creation step comprises the steps of:

a) Perform a first pass scan using the component labeled image and component characteristics to create a first pass intermediate adaptive distance image and an adaptive shortest distance component label image wherein the adaptive distance image having distance metric based on the component weighted length;

b) Perform a second pass scan using the first pass intermediate adaptive distance image and the adaptive shortest distance component label image to create an adaptive distance transform image and an updated adaptive shortest distance component label image wherein the adaptive distance transform image having distance metric based on the component weighted length.

3. An adaptive cell segmentation method using a computer or processor to carry out the steps of:

a) Input a nuclei mask image, its component labeled image and component characteristics;

b) Input a cell mask image;

c) Perform adaptive nuclei region partition using the nuclei mask component labeled image and component characteristics to create adaptive nuclei mask ZOI using a length function 1 wherein a component specific weighting factor w is applied to the length function to generate a component weighted length;

l[(i,j),Z(i,j)]=l(i,j)*w(Z(i,j))

wherein w is a general function of component characteristics that is related to one of component size, component confidence, or types of components and Z(i,j) is component labeled image value at image position (i,j);

d) Perform adaptive cell region separation using the cell masks and the adaptive nuclei mask ZOI to generate adaptive cell separated regions wherein the adaptive cell mask having the same component label in the nuclei mask ZOIs is considered as one cell region.