Image processing improvements or improvements related to image processing
A cross-correlation and blood vessel image technology, applied in image data processing, image analysis, image enhancement, etc., can solve the problem of not being able to find enough
Active Publication Date: 2018-05-04
OPTOS PLC
View PDF2 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the inventors believe that the technique proposed in EP'988 will not be able to find enough corner features in the vasculature in a typical retinal image to achieve reliable image matching and registration, especially to achieve Image matching and registration between images of
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0072] figure 1 is a flowchart detailing the method steps of a registration algorithm for affine registration of retinal image pairs or image sequences.
[0073] figure 1 Method steps for processing digital retinal image data are shown. The first image may be referred to as a "base image" and each subsequent image may be referred to as an "input image".
[0074] refer to figure 1 , 2a and 2b, the first step 100 of the method is providing first and second digital retinal image data. The first and second digital retinal image data are represented by first and second digital retinal images 10a, 10b. The first and second digital vessel image data include incident illumination intensities on one or more pixels used to generate the image data. The first and second digital retinal images 10a, 10b illustrated here were obtained by a widefield scanning laser ophthalmoscope (SLO) well known in the art. Such as Figure 2a and 2b As shown, the first and second retinal images 10 a...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
An image processing device that uses first and second digital vessel image data to register two images. The two images can be from different imaging modalities. The first and second images are processed through a two-dimensional directional filter (500) which has the effect of generating clusters of orthogonally adjacent image data points, with each orthogonally adjacent image data point in the cluster between The magnitude of the intensity gradient of is less than a predetermined value. Common clusters between the first and second image data are then identified using a corner detection algorithm (600). The directional filter produces "stepped" features where vessel features would otherwise appear to have smooth edges. These numerous features are identified by a corner detection algorithm and can be used for registration of common clusters between the first and second image data (1000). The filter may be a rotated Gabor filter that matches vessel features in the image.
Description
technical field [0001] The present invention relates to image processing or improvements related to image processing and in particular, but not exclusively, to a method and apparatus for registering pairs or sequences of images of the vasculature such as retinal images. Background technique [0002] Imaging systems, such as scanning laser ophthalmoscopy (SLO), are known systems that use one or more digital image sensors to capture retinal image data. A digital image sensor for SLO is typically a single sensor where the light intensity signal is synchronized with the scan position signal to generate a single data stream that can be synchronized to a 2D image. The digital image sensor may optionally include an array of light-sensitive picture elements (pixels). A retinal image generated by an SLO or other retinal imaging device such as a fundus camera is typically a two-dimensional array of pixels and is referred to as a digital retinal image. [0003] The set of intensity va...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/13G06T5/00
CPCG06T2207/30041G06T2207/20164G06T2207/30101A61B3/0025A61B3/1225G06T7/32G06T7/337G06T7/136G06T5/20
Inventor 大卫·克利夫顿拉尔夫·艾伦·皮诺克
Owner OPTOS PLC