Recent Patents in Signal Processing (February 2017) – Active contours

For our February 2017 issue, we cover recent patents granted in the area of the analysis and application of active contours.

Patent no. 9,495,756 presents a system and method for active contour segmentation where an image for a structure and an initial position on the image are received, a multi-scale image representation comprising successive image levels each having associated therewith a representation of the image is computed, a representation of the image at a given level having a different image resolution than that of a representation of the image at a subsequent level, a given one of the levels at which noise in the image is removed is identified, the initial position is set as a current contour and the given level as a current level, the current contour is deformed at the current level to expand into an expanded contour matching a shape of the structure, the expanded contour is set as the current contour and the subsequent level as the current level, and the steps are repeated until the last level is reached.

In patent no. 9,142,184 a device receives an image for an active contour model, where the active contour model includes a first module and a second module. The device receives an initial contour for the image and the active contour model, and executes a speed function of the first module, based on the image and the initial contour, to generate a speed function result. The device executes a numerical minimization method of the second module, based on the image and the speed function result, to generate a final contour, and provides the image and the final contour for display.

As descibed in patent no. 9,076,223 a device is configured to determine a contour vector that delineates an object in an image from a remaining portion of the image, and to generate a hash value by applying a hash function to the contour vector. The device is configured to compare the hash value to a previous hash value generated by applying the hash function to a previous contour vector, where the previous contour vector is determined prior to determining the contour vector. The device is configured to determine that the hash value matches the previous hash value and, based on determining that the hash value matches the previous hash value, segment the image using the contour vector.

In patent no. 8,942,441 a computer-implemented segmentation method is used to process an image representing a plurality of nuclei. The method is implemented in a computer having a processor and a physical memory. A set of instructions are provided to the processor the physical memory of the computer. The processor is configured by executing the set of instructions in the physical memory so as to automatically segment the image by: thresholding a grey-scale image to create a black and white image; identifying objects in the black and white image and removing objects failing to meet predetermined criteria; extracting objects; and applying an edge detector on the segmented image to identify the edges of the nuclei. Overlapping nuclei are split to improve results.

Patent no. 8,811,679 introduces a method for realigning network elements in a digital map based on a new data set using active contour manipulation techniques. The active contour optimization process is controlled so that certain desirable features in the original shape of the active contour(s) are not distorted. In the case of network features which may be represented by two or more open active contours, the active contours are coupled prior to optimizing so that the original desirable shape, e.g., the parallel spacing of lanes in a dual carriageway road, may be maintained during the optimization process. In the case of certain closed active contours, for example roundabout geometries, a single closed active contour may be coupled to itself by linking its control points one-to-another and maintaining a proportional direction and distance movement during the optimizing process. As a result, data base networks can be assessed and aligned with less difficulty and geometric distortion.

A method of image object tracking and segmentation is provided in patent no. 8,781,160. The method includes defining an initial contour for tracking an image object and partitioning the initial contour into a plurality of contour segments. The method also includes estimating a weighted length of each of the plurality of contour segments and generating a desired contour by converging the plurality of contour segments to a plurality of edges of the image object using the estimated weighted length.

Certain embodiments in patent no. 7,831,104 provide systems and methods for using geometric wavelets based on active contours for signal processing. Certain embodiments may allow image processing. A system includes providing an image processing circuitry that computes a segmentation tree using active contours for an input image data; creates a geometric wavelets representation using the segmentation tree; and then generate an image based on geometric wavelet sparse representation extracted from the geometric wavelets representation. The geometric wavelet sparse representation may comprise M most active geometric wavelets from a set of N geometric wavelets in the geometric wavelets representation. The image processing circuitry may recursively find sub-domains and multivariate polynomials for each domain at each stage in the computing of the segmentation tree. Each of the sub-domains is a domain for a succeeding stage for the segmentation tree. The recursively finding may be terminated, for example, when each domain at a stage comprises less than a determined number of pixels.

An exemplary method of detecting a structure in a three-dimensional (3D) computed tomography ("CT") knee image is provided in patent no. 7,620,226. An intensity prior mask is created based on a two-dimensional (2D) slice of the CT image. A seed point in the 2D slice is received or estimated. The seed point is in a Hoffa's pad region. A geometric prior mask is created based on the seed point. The intensity prior mask and the geometric mask are combined to form a region mask. A level-set contour is segmented from the 2D slice. The step of segmenting is constrained by the region mask.

If you have an interesting patent to share when we next feature patents related to active contours, or if you are especially interested in a signal processing research field that you would like to highlight in this section, please send email to Csaba Benedek (benedek.csaba AT sztaki DOT mta DOT hu).

References

Number: 9,495,756
Title: Multi-scale active contour segmentation
Inventors: Rivet-Sabourin; Geoffroy (Stoneham, CA)
Issued:  November 15, 2016
Assignee: Laboratoires Bodycad INC. (Quebec, Quebec, CA)

Number: 9,142,184
Title: Speed function module and numerical minimization module for an active contour model
Inventors: Tiwari; Spandan (Framingham, MA), Eddins; Steven L. (Milford, MA), Mather; Jeff D. (Milford, MA), Taylor; Alexander J. (Jamaica Plain, MA), Uthama; Ashish (Arlington, MA), Yale; Mara M. (Framingham, MA)
Issued:  September 22, 2015
Assignee: The MathWorks, Inc. (Natick, MA)

Number: 9,076,223
Title: Fast stopping criterion for active contour algorithms
Inventors: Tiwari; Spandan (Framingham, MA), Eddins; Steven L. (Milford, MA)
Issued:  July 7, 2015
Assignee: The MathWorks, Inc. (Natick, MA)

Number: 8,942,441
Title: Optimizing the initialization and convergence of active contours for segmentation of cell nuclei in histological sections
Inventors: Nielsen; Birgitte (Oslo, NO), Maddison; John (Sussex, GB), Danielsen; Havard (Oslo, NO)
Issued:  January 27, 2015
Assignee: Institute for Medical Informatics (Oslo, NO)

Number: 8,811,679
Title: Digital map editing process using active contour manipulation
Inventors: Bekaert; Tim (Heule, BE), Clarysse; Pascal (Koksijde, BE)
Issued:  August 19, 2014
Assignee: TomTom Belgium N.V. (Gent, BE)

Number: 8,781,160
Title: Image object tracking and segmentation using active contours
Inventors: Viswanathan; Srikrishnan (Mumbai, IN), Chaudhuri; Subhasis (Mumbai, IN)
Issued:  July 15, 2014
Assignee: Indian Institute of Technology Bombay (Powai, Mumbai, IN)

Number: 7,831,104
Title: Using geometric wavelets based on active contours for signal processing
Inventors: Dekel; Shai (Ramat-Hasharon, IL), Sherman; Alexander (Modiin, IL)
Issued:  November 9, 2010
Assignee: General Electric Company (Schenectady, NY)

Number: 7,620,226
Title: Semi-local active contour for a faint region detection in knee CT images
Inventors: Unal; Gozde (West Windsor, NJ), Slabaugh; Gregory G. (Princeton, NJ), Shen; Hong (Plainsboro, NJ)
Issued:  November 17, 2009
Assignee: Siemens Medical Solutions USA, Inc. (Malvern, PA)