GB2466069A - Method of identifying the texture of portions of a digital image by analysing a composite texture profile - Google Patents
Method of identifying the texture of portions of a digital image by analysing a composite texture profile Download PDFInfo
- Publication number
- GB2466069A GB2466069A GB0822715A GB0822715A GB2466069A GB 2466069 A GB2466069 A GB 2466069A GB 0822715 A GB0822715 A GB 0822715A GB 0822715 A GB0822715 A GB 0822715A GB 2466069 A GB2466069 A GB 2466069A
- Authority
- GB
- United Kingdom
- Prior art keywords
- texture
- profile
- values
- pixel
- composite
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000002131 composite material Substances 0.000 title claims abstract description 103
- 238000000034 method Methods 0.000 title claims abstract description 92
- 210000004872 soft tissue Anatomy 0.000 claims abstract description 4
- 238000012545 processing Methods 0.000 claims description 5
- 230000001419 dependent effect Effects 0.000 claims 6
- 238000002595 magnetic resonance imaging Methods 0.000 abstract description 8
- 238000012512 characterization method Methods 0.000 abstract description 5
- 210000002307 prostate Anatomy 0.000 abstract description 5
- 238000002059 diagnostic imaging Methods 0.000 abstract description 4
- 230000002159 abnormal effect Effects 0.000 abstract description 2
- 238000002604 ultrasonography Methods 0.000 abstract description 2
- 239000003814 drug Substances 0.000 abstract 1
- 238000003325 tomography Methods 0.000 abstract 1
- 210000001519 tissue Anatomy 0.000 description 5
- 238000003491 array Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012935 Averaging Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 241001465754 Metazoa Species 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000002591 computed tomography Methods 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 238000002600 positron emission tomography Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000000275 quality assurance Methods 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- MYLBTCQBKAKUTJ-UHFFFAOYSA-N 7-methyl-6,8-bis(methylsulfanyl)pyrrolo[1,2-a]pyrazine Chemical compound C1=CN=CC2=C(SC)C(C)=C(SC)N21 MYLBTCQBKAKUTJ-UHFFFAOYSA-N 0.000 description 1
- 241000238631 Hexapoda Species 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 244000144992 flock Species 0.000 description 1
- 244000144980 herd Species 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 150000002739 metals Chemical class 0.000 description 1
- 238000001000 micrograph Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
- G06T7/41—Analysis of texture based on statistical description of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Probability & Statistics with Applications (AREA)
- Image Processing (AREA)
Abstract
A method of creating an electronic library for characterising texture types of portions of digital images comprising pixels with a texture value for each respective position, a texture profile that identifies pixels within the portion that have texture values within a particular range, and then combining a plurality of related texture profiles to create and store a composite texture profile characterising the distribution of texture values within the range and an indication of the particular texture type. The composite texture profile may be a summation, normalised summation, average, or average of a cluster of the texture profiles. Binary representation may be used to distinguish different ranges of texture values. Texture characterisation is used in medicine to assist in identifying abnormal growth in soft tissue, such as cancerous growth in the prostate gland, and the digital images may comprise medical imaging data, such as magnetic resonance imaging (MRI), tomography or ultrasound data.
Description
TEXTURE CHARACTERISATION
Field
The present invention relates to texture characterisation. In particular, the present invention relates to characterising texture types of digital images, for example digital images omprising medical imaging data (e.g. magnetic resonance imaging (MRI) data). Texture characterisation is important in a range of applications including the characterisation of tissue texture to assist in identifying abnormal growth such as cancerous growth, for example in the prostate gland.
Background
One known technique to characterise texture type is for a specialist such as a radiologist to analyse an image such as an M image.
The present invention seeks to provide a more automated technique.
Summary
One aspect of the present invention provides a method for creating an electronic library for characterising texture types of portions of digital images, a portion of a digital image comprising a plurality of pixels, each pixel having a texture value which is representative of texture at a respective position, wherein a portion of the digital image represents a particular texture type and a texture profile for a portion identifies pixels of the portion having texture values within a particular range of texture values, the method comprising: combining a plurality of texture profiles for portions of a particular texture type to create a composite texture profile for a representative portion of the digital image having the particular texture type, wherein the composite texture profile characterises the distribution, across the pixels of the representative portion, of texture values within the particular range; and storing in the electronic library the composite texture profile and an associated indication of the particular texture type.
Another aspect of the present invention provides a method of identifying the particular texture types of pixels within a digital image, the digital image comprising a plurality of pixels, each pixel having a texture value which is representative of texture at a respective position, the method using an electronic library which stores for a plurality of particular texture types a composite texture profile for a representative portion of a digital image and an associated indication of the particular texture type, wherein a composite texture profile characterises the distribution, across the pixels of the representative portion, of texture values within a particular range, the method comprising processing the digital image to allocate a texture type to a pixel of the digital image by: combining a plurality of texture profiles associated with the pixel, each texture profile identifying pixels of a portion having texture values within a particular range of texture values, to create a generated composite texture profile for a portion of the digital image around the pixel, wherein the composite texture profile characterises the distribution, across the pixels of the portion of the digital image around the pixel, of texture values within the particular range; comparing the generated composite texture profile with one or more composite texture profiles from the electronic library to identify a best fitting composite texture profile; and allocating the texture type associated with the best fitting composite texture profile to the texture type of the pixel.
In embodiments of the present invention digital images may comprise medical data/medical imaging data, and optionally medical data of soft tissue such as magnetic resonance imaging data.
Another aspect of the present invention provides a system comprising a menioiy and a processor, wherein the processor is arranged to perform one or more of the above methods.
Another aspect of the present invention provides a computer-readable medium having computer-executable instructions adapted to cause a computer system to perform one or more of the above methods.
Accordingly, a more automated technique is provided.
Other aspects and features of the present invention will be appreciated from the following
description and the accompanying claims.
Brief description of the drawings
Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings in which like reference numerals are used to depict like parts. In the drawings: Figure 1 is a flow diagram illustrating a method in accordance with an embodiment of the invention; Figures 2-4 illustrate digital images; Figure 5 and 6 illustrate texture profiles generated for portions of a digital image; Figures 7 and 8 illustrate the generation of composite texture profiles; Figure 9 is a flow diagram illustrating a method in accordance with another embodiment of the invention; Figure 10 illustrates clustering to generate composite texture profiles; Figure 11 is a flow diagram illustrating a method in accordance with another embodiment of the invention; Figure 12 illustrates a technique to compare generated composite texture profiles with those from an electronic library; and Figure 13 illustrates a digital image which has had its pixels classified by texture type.
Detailed description of embodiment(s)
Configuring the electronic library A method is provided for creating an electronic library for characterising the texture types of portions of digital images, for example M images of prostates. Such a method will be described with reference to Figure 1, but to assist with understanding the steps of the method of Figure 1, a description of Figures 2 to 8 will first be given.
A
Figure 2 illustrates a digital image 20. The digital image comprises a plurality of pixels, and each pixel has a texture value which is representative of the texture of the article at a respectivc position. For simplicity, not all pixels of the digital image 20 are illustrated. Some of the pixels of the digital image are illustrated as a 20x15 grid, and within this grid three example pixels 22, 24 and 26 are shown. Each pixel is represented by a location in the digital image and by a single or a plurality of values (representing, for example, grey-level, colour, etc), which will be referred to as the texture value. A texture value for a pixel is a numerical representation of the texture of the article at the position of the location represented by the pixel, for example any numerical value between 0 and 255.
The digital image may comprise medical imaging data, for example soft tissue discrimination imaging data such as magnetic resonance imaging (MRI) data. MRI data is generated by an MIII scanner and the data typically forms a set of digital images which together form 3D/volumetric data.
Figure 3 also illustrates the digital image 20 and example pixels 22, 24, 26. Around each pixel a respective portion 32, 34, 36 of the image is illustrated. Typically, a portion is a subset of the digital image. Portion 32 is a 3x3 array of pixels around and including pixel 22.
Portion 36 is a 5x5 array of pixels around and including pixel 26. Portion 34 is an irregular collection of 12 cormected pixels around and including pixel 34. In different embodiments, different shaped and sized portions may be used, although portions in the form of 5x5 arrays such as portion 36 will be used when describing this feature with reference to later figures.
A set of predetermined particular ranges of texture values can be defined. Figure 4 shows a digital image 20 in which the texture value of each pixel falls into one of four distinct ranges.
For example, the pixels marked with vertical shading such as pixel 45 each have a value within a first range; the pixels marked with diagonal top left to bottom right shading such as pixel 26 each have a value within a second range; the pixels marked with horizontal shading such as pixel 47 each have a value within a third range; and the pixels marked with diagonal bottom left to top right shading such as pixel 48 each have a value in a fourth range. For example, if the texture values can have any possible value between 0 and 255, the first range could be between 0 and 63, the second range between 64 and 127, the third range between 128 and 191 and the fourth range between 192 and 255. In this example, texture values of say 7, 9 27, or 36 would all fall in the first range and so on.
A digital image may comprise data representing one or more particular texture types. In Figure 4 two texture types 42 and 44 are illustrated. For the method of creating the electronic library the particular texture types would be determined and classified by a texture classification expert such as a radiologist where the image is a medical image. The radiologist would classify the texture types as "nodular tissue" or "radiolucent tissue" for
example.
A texture profile for a portion of the digital image can be generated. A texture profile identifies pixels of the respective portion having texture values within the particular range of texture values such as the ranges of texture values already described in relation to Figure 4.
Figure 5 illustrates the part of the digital image 20 having the two texture types 42 and 44.
Four 5x5 texture profiles 52, 54, 56 and 58 around four pixels 12, 14, 16, 18 are shown. Each texture profile is a binary representation of the 25 pixels around the relevant pixel, which in this example is in the centre of the 5x5 array. Each texture profile comprises cells or elements which each have a value which is in the same range as the value for the centre cell. These cells are represented as having a first binary value, here "1 ". The values in the texture profile of cells which fall outside this particular range are given a different binary value. This value is not depicted here but can be considered as "0". Note, with reference to the description already given with reference to Figure 4, pixels which have the same shading have values falling within the same range of texture values.
Instead of a set of predetermined particular ranges of texture values being defined, which can be considered as an absolute way of setting the ranges, whether or not a texture value falls within a particular range can be determined with reference to a relative range around the texture value for a particular pixel. That is, the particular range of texture values for the texture profile can be defined as being a relative range (+1-a particular value) around the texture value of the pixel around which the portion is positioned. Describing this further with reference to Figure 6, the figure shows a digital image in which each pixel has a texture value of any value between 0 and 255. Typical +1-values within such a total range are +1-16 or +1- 32. For simplicity, in Figure 6 the same shading as Figure 5 has been used but it should be noted that there are more than four values across the image; that is, the pixels may have any value between 0 and 255 but the complete extent of these grey levels are not depicted in the figure. Referring to the texture profile 62 of Figure 6 which is created around the central pixel, the values in the image for the 5x5 texture profile may be: 55, 89, 44, 45, 96 92, 51, 33, 99, 29 29, 35, 85, 35, 37 28, 87, 37, 38, 91 42, 42, 22, 86, 32 Any values which are +1-16 of the central value 85 can be allocated a "1" and all other values outside this range a "0", making: 0, 1, 0, 0, 1 1, 0, 0, 1, 0 0, 0, 1, 0, 0 0, 1, 0, 0, 1 0, 0, 0, 1, 0 This texture profile is depicted as texture profile 62 in Figure 6.
For a single texture profile, whether determined in the way described with reference to Figure or that of Figure 6, further texture profiles can be generated for associated portions of the digital image 20. For example, texture profiles for portions having the same particular texture type as the single texture profile can be determined.
In more detail, a first texture profile for a portion of a digital image around a pixel having a value within the particular range of texture values can be generated, and one or more other texture profiles for one or more other portions can be generated. Each of the other portions can, for example, be around a respective other pixel within the first texture profile having a value within the range. With reference to Figure 7, these one or more other texture profiles can be generated for portions positioned around a respective one or more pixels within the first texture profile which each (i) has a binary value equal to the first binary value and (ii) is connected, either directly or through one or more other pixels having the first binary value, to the pixel around which the first portion is positioned. In the example illustrated in Figure 7, the first texture profile 62 has four pixels which meet these criteria (the four pixels that are not the centre pixel but which have a "1" in them and fall along the diagonal sequence of pixels from the bottom left to the top right of the profile 62). As shown in Figure 7, four other texture profiles around these other pixels can be generated. These four texture profiles and the first texture profile 62 can be combined (in the illustrated example by adding) to produce the composite texture profile 72. This composite texture profile is for a representative portion of the digital image having the particular texture type and it characterises the distribution, across the pixels of the representative portion (here a 5x5 portion), of texture values within the particular range used to produce the texture profiles.
The composite texture profile 72 is a summation of the plurality of texture profiles. In the illustrated example the plurality consists of five texture profiles. Such a composite texture profile can be normalised to produce a normalised summation of the plurality of texture profiles. A composite texture profile 74 in the form of a normalised summation of the plurality of texture profiles is also depicted in Figure 7. The entries in the values of the pixels of composite texture profile 74 add up to one (rounding makes it appear from the figure that the total is 0.9, but each entry is in fact 0.1111 making the total 1).
It is possible to generate multiple composite texture profiles for a particular texture type by performing the process described with reference to Figure 7 but starting with another pixel within the digital image. These multiple texture profiles can be averaged. A composite texture profile 82 in the form of an average of multiple composite texture profiles is illustrated in Figure 8.
Irrespective of the actual form, each composite texture profile 72, 74, 82 characterises the distribution, across the pixels of a representative portion of the texture type, of texture values within a particular range of texture values. A composite texture profile can comprise an array having a plurality of cells or elements corresponding to the plurality of pixels of the portion of the digital image for which the texture profile is generated. Each element can comprise an entry representing the likelihood, when compared with the other elements, of a texture value within the particular range being within that element for a representative portion of the particular texture type. The composite texture profile can be considered as being a probability map which, in one impiementation, has values in its elements showing the likelihood, when compared with the other elements, of a texture value within the particular range being within each element for a representative portion of the particular texture type. The values in these elements may sum to one across the whole composite texture profile.
It will be appreciated that the composite texture profile is generated by combining a plurality of texture profiles. In other words, the composite texture profile is generated based on, or in dependence on, the plurality of texture profiles.
Referring now back to Figure 1, a flow diagram of a method for creating an electronic library for characterising the texture types of portions of digital images is shown. Typically, the steps S2 to S6 are performed for some or all of the pixels within a digital image which have been ascribed a particular texture type.
Referring to step S2, for a single pixel within the digital image in an area which has been identified as being of a particular texture type (e.g. area 42 of Figure 4) a portion around the single pixel is defined (e.g. a 5x5 portion around pixel 12 of Figure 5).
Referring to step S4, for the portion around the pixel a first texture profile is generated (e.g. texture profile 52 of Figure 5).
Referring to step S6, a composite texture profile is generated, for example by performing the process as described with reference to Figure 7 to generate the composite texture profile 72.
The process of steps S2 to S6 is typically performed for more than one pixel, for example with reference to Figure 5 these steps can be performed for all pixels falling within area 42 (i.e. all pixels having a particular texture type) and having a texture value in a particular range (e.g. all pixels depicted in Figure 5 with vertical shading). That is, the process can be performed for all pixels in a first of the ranges.
At step S8, the composite texture profile can optionally be processed further, for example to produce composite texture profiles 74 or 82 which have been described with reference to Figures 7 and 8. The composite texture profile is then stored in the electronic library together with an indication of the particular texture type.
The process of steps S2 to S8 can be performed for each pixel in each range, for example for all pixels in the area of the digital image denoted as texture type 42 in Figure 5. That is, after having performed the process for all pixels in a first of the ranges, the process can be performed for all pixels in the next range and so on until all ranges and therefore all pixels have been processed.
Figure 9 shows a variation of the method of Figure 1. In the Figure 9 method, steps S2 to S6 are the same as those of Figure 1, except that in Figure 9 the particular range of texture values for the texture profile generated in step S4 can be defined as being a relative range of values around the texture value of the pixel, as described with reference to Figure 6.
In contrast to the process of Figure 1, in step SlO of Figure 9 a composite texture profile in the form of an average of a cluster of multiple composite texture profiles for the particular texture type is generated. In the Figure 9 method, a plurality of such composite texture profiles for a respective plurality of multiple composite texture profiles is stored.
In more detail, Figure 10 shows how clusters 116 of multiple composite texture profiles can be generated. Multiple composite texture profiles 74 are generated, as already described.
Instead of averaging these (as described with reference to Figure 8) subsets of these multiple composite texture profiles can be grouped together using a clustering technique which groups together items having one or more similar characteristics. Clustering techniques are known and one suitable known clustering technique that can be used is the kMeans technique.
Figure 10 illustrates a subset 112 of composite texture profiles being grouped together, for example by averaging, to form one cluster 116 of multiple composite texture profiles and another subset 114 of composite texture profiles being grouped together, for example by averaging, to form another cluster 116 of multiple composite texture profiles.
The processes described with reference to Figures 1 or 10 can be performed for a plurality of test articles, i.e. articles with particular, known texture types, such as MRI test images of prostates, each having an ascribed texture type. Accordingly, an electronic library can be generated which stores for a plurality of particular texture types a composite texture profile for a portion of a digital image and an associated indication of the particular texture type.
Using the electronic library to identify texture type A method is provided of identifying texture types of pixels of a digital image, for example an MRI image of a prostate of a patient. Accordingly, an automated identification tool for identifying particular texture type within an image, for example cancerous cells, can be provided.
The method of identifying the texture type of pixels of a digital image has similar features to the method of configuring the electronic library, with the notable exception that whereas to configure the electronic library the texture type of the relevant part of the digital image was known, when identifying the texture type of the pixels within a digital image this characteristic is initially unknown and the electronic library is used to identify it. For simplicity, features of the earlier figures will be referred to when describing using the electronic library to identify the texture type, although it should be noted that in this section the texture type(s) of the pixels of the digital image are initially unknown and then determined using the data in the electronic/texture library.
The digital image which is to have the texture type of its pixels classified comprises a plurality of pixels, each pixel having a texture value which is representative of the texture of the article at a respective position (see for example the digital image 20 of Figure 2).
In overview, to identify the texture type of a pixel a composite texture profile is generated for a pixel of a digital image. This generated composite texture profile characterises the distribution, across the pixels of the portion of the digital image around the pixel, of texture values within a particular range. The generated composite texture profile is compared with one or more, or typically all, composite texture profiles from the electronic library to identify a best fitting composite texture profile from the library. The texture type associated with the best fitting composite texture profile is allocated to the texture type of the pixel. This process is typically repeated for each pixel in the digital image.
Referring to Figure 11, a method of identifying the particular texture type of a pixel within a digital image is shown. At step S22 a portion around the pixel is identified (see e.g. the portion 36 around pixel 26 of Figure 3). Then, at step S24 a texture profile for this portion is generated (see e.g. texture profile 52 of Figure 5). At step S26, a composite texture profile is generated, for example by performing the process as described with reference to Figure 7 to generate the composite texture profile 72. These steps of the process are the same as the first three steps of Figures 1 and 9.
At step S28 the generated composite texture profile is compared with the composite texture profiles from the electronic library. A best fitting composite texture profile from the library is identified. The texture type associated with the best fitting composite texture profile is allocated to the texture type of the pixel.
The process of Figure 11 will typically be performed for each pixel of interest in the digital image; often this will be for each pixel in the image. For example, a plurality of predefined ranges can be defined and for each range a binary representation of the whole image can be generated, in which a pixel of the digital image is given a first binary value if its texture value is within the range and in which pixels having texture values outside the range are given a second binary value. Each range can be processed in turn. For a binary representation for a first range, a portion can be positioned around a pixel having a binary value equal to the first binary value, and steps S22 to S28 performed. This process can be performed for each pixel having the first binary value. The same process can be used for a binary representation for each of the other ranges in turn until each and every pixel has been given a texture type.
Alternatively, each pixel can be processed in turn using a relative ranging technique as previously described with reference to Figure 6.
A representation of the digital image which has had its pixels classified using the described method is shown in Figure 13. Each pixel has been allocated a texture type (here depicted by different shading). The pixels in a first area 262 have each been allocated one texture type and the pixels in another area 264 have been allocated another texture type. The pixels between the two areas have been allocated one or other of the texture types allocated to the picis of areas 262 or 264. In panicuiar implementations, each texture type can be depicted with a different colour and the shapes of the areas will, of course, not necessarily be rectangular. For example, various areas of a digital image of a tissue may be irregular shapes.
Typically, the portion of the digital image for which a texture profile is generated is of a predetermined size which is the same size as the size of the composite texture profiles stored in the electronic library.
Referring back to step S28 of Figure 11, comparing the generated composite texture profile with a composite texture profile from the electronic library to identify the best fitting composite texture profile can use a summed comparison value and comprise the following steps (although other known comparison techniques such as the Euclidean distance technique can be used).
First, as an optional pre-processing step to reduce the overall amount of processing that is needed, the mutual minimum values of each profile can be subtracted from each pixel of the two profiles and the resulting profiles can be used for the calculation of the comparison values. Alternatively, the unprocessed profiles can be used for the following steps.
For each possible combination of a pixel in the generated composite texture profile (either with or without the above pre-processing) a pixel in the composite texture profile from the library, a comparison value is calculated by multiplying the two respective pixel values by each other and the distance between the two pixels.
Then, the comparison value for each pixel combination is summed to determine a summed comparison value.
The process is typically performed for the generated profile and each composite texture profile from the library. The composite texture profile from the library with the lowest summed comparison value is selected or identified as the best fitting standardised texture profile.
Figure 12 illustrates the comparison step (S28 of Figure 11) used to compare the two composite texture profiles in more detail.
Referring to Figure 12, a digital image similarity measure (DISM) is calculated. Small 2x2 composite texture profiles 242 and 244 are used for illustrative purposes in Figure 12.
Composite texture profile 242 is the generated composite texture profile and composite texture profile 244 is from the electronic library. Array 246 comprises the minimum values of arrays 242 and 244. The minimum values held in this array are subtracted from the arrays 242 and 244 to create arrays 243 and 245 respectively. The summation shown is calculated to obtain the DISM. The distance of between the elements in the array used in one implementation is the manhattan distance, but alternatives can be used.
In one or more implementations the electronic library is a database having records, each record comprising: a composite texture profile as an array; and an indicator of texture type.
A system or computer such as a general-purpose computer can be configured or adapted to perform the described methods. In one embodiment the system comprises a processor, memory, and a display. Typically, these are connected to a central bus structure, the display being connected via a display adapter. The system can also comprise one or more input devices (such as a mouse and/or keyboard) and/or a communications adapter for connecting the computer to other computers or networks. These are also typically connected to the central bus structure, the input device being connected via an input device adapter.
In operation the processor can execute computer-executable instructions held in the memory and the results of the processing are displayed to a user on the display. User inputs for controlling the operation of the computer may be received via input device(s).
A computer readable medium (e.g. a carrier disk or carrier signal) having computer-executable instructions adapted to cause a computer to perform the described methods may be provided.
Embodiments of the invention have been described by way of example only. It will be appreciated that variations of the described embodiments may be made which are still within the scope of the invention.
For example, although not a limited list, the techniques can be used to analyse texture in images in the fields of medical imagery, materials science, food science, and satellite, airborne and other remote sensing imagery. In medical imagery, CT (computed tomography), MRI, Ultrasound, PET (positron emission tomography), microscopy images of human or animal tissues, cells, bones or nerves can be processed. In materials science, images of materials such as manufactured fabrics or metals, or raw materials can be processed for quality assurance purposes. Similarly, in food science, images of finished products or raw materials can be processed for quality assurance purposes. In satellite, airborne and other remote sensing imagery, images can be processed for automated and knowledge based mapping of weather patterns, geographical features (e.g. wet lands/drought), topological and geological structures (e.g. dunes/coastline/riverbeds/mountains), plants (e.g. crop types/woodland density and edges), animals (e.g. insect swarms/herds/flocks/classification), and man-made features (e.g. traffic congestion/road boundaries). It will be appreciated that in each case a different article or articles are scanned. Typically, the digital image is therefore a scanned digital image or, more particularly, a scanned digital image of a real world article or articles (or entity/entities or object/objects).
Claims (30)
- Claims 1. A method for creating an electronic library fOr characterising texture types of portions of digital images, a portion of a digital image comprising a plurality of pixels, each pixel having a texture value which is representative of texture at a respective position, wherein a portion of the digital image represents a particular texture type and a texture profile for a portion identifies pixels of the portion having texture values within a particular range of texture values, the method comprising: combining a plurality of texture profiles for portions of a particular texture type to create a composite texture profile for a representative portion of the digital image having the particular texture type, wherein the composite texture profile characterises the distribution, across the pixels of the representative portion, of texture values within the particular range; and storing in the electronic library the composite texture profile and an associated indication of the particular texture type.
- 2. A method according to claim 1, comprising: generating a first texture profile for a portion of a digital image around a pixel having a value within the particular range of texture values; and generating one or more other texture profiles for one or more other portions, each of the other portions being around a respective other pixel within the first texture profile having a value within the range, wherein the first texture profile and the one or more other texture profiles form the plurality of texture profiles which are combined.
- 3. A method according to claim 1 or 2, wherein the composite texture profile stored in the electronic library is a summation of the plurality of texture profiles.
- 4. A method according to any preceding claim, wherein the composite texture profile stored in the electronic library is a normalised summation of the plurality of texture profiles.
- 5. A method according to any preceding claim, wherein the composite texture profile stored in the electronic library is an average of multiple composite texture profiles for the particular texture type.
- 6. A method according to any preceding claim, wherein the composite texture profile stored in the electronic library in an average of a cluster of multiple composite texture profiles for the particular texture type, and wherein a plurality of such composite texture profiles for a respective plurality of multiple composite texture profiles is stored.
- 7. A method according to any preceding claim, wherein the texture profiles each identify pixels of the portion having texture values within a particular range using a binary representation, wherein the binary representation uses binary values to distinguish, as a first of two binary values, the pixels of the portion having texture values within the particular range, from as a second of two binary values, the pixels of the image having texture values outside the particular range.
- 8. A method according to claim 7, when dependent on claim 2, comprising determining the position of the portion for the first texture profile by positioning the portion around a pixel having a binary value equal to the first binary value.
- 9. A method according to claim 8, when dependent on claim 2, comprising generating the one or more other texture profiles for portions positioned around a respective one or more pixels within the first texture profile having a binary value equal to the first binary value.
- 10. A method according to claim 8, when dependent on claim 2, comprising generating the one or more other texture profiles for portions positioned around a respective one or more pixels within the first texture profile which each (i) has a binary value equal to the first binary value and (ii) is connected, either directly or through one or more other pixels having the first binary value, to the pixel around which the first portion is positioned.
- 11. A method according to any preceding claim, wherein a set of predetermined particular ranges of texture values is defined and the method is performed for different particular ranges.
- 12. A method according to any of claims ito 10, wherein the particular range of texture values for the texture profile is defined as being a relative range of values around the texture value of the pixel around which the portion is positioned.
- 13. A method according to claim 11 or 12, wherein the method is performed for a portion around each pixel of the digital image.
- 14. A method according to any preceding claim, wherein the portion of the digital image for which a texture profile is generated is of a predetermined size.
- 15. A method of identifying the particular texture types of pixels within a digital image, the digital image comprising a plurality of pixels, each pixel having a texture value which is representative of texture at a respective position, the method using an electronic library which stores for a plurality of particular texture types a composite texture profile for a representative portion of a digital image and an associated indication of the particular texture type, wherein a composite texture profile characterises the distribution, across the pixels of the representative portion, of texture values within a particular range, the method comprising processing the digital image to allocate a texture type to a pixel of the digital image by: combining a plurality of texture profiles associated with the pixel, each texture profile identifying pixels of a portion having texture values within a particular range of texture values, to create a generated composite texture profile for a portion of the digital image around the pixel, wherein the composite texture profile characterises the distribution, across the pixels of the portion of the digital image around the pixel, of texture values within the particular range; comparing the generated composite texture profile with one or more composite texture profiles from the electronic library to identify a best fitting composite texture profile; and allocating the texture type associated with the best fitting composite texture profile to the texture type of the pixel.
- 16. A method according to claim 15, comprising: processing the digital image to allocate a texture type to a pixel of the digital image by: generating a first texture profile for a portion of a digital image around a pixel having a value within a particular range of texture values, wherein the texture profile identifies pixels of the portion having texture values within the particular range; and generating one or more other texture profiles for one or more other portions, each of the other portions being around a respective other pixel within the first texture profile having a value within the range, wherein the first texture profile and the one or more other texture profiles form the plurality of texture profiles for combining.
- 17. A method according to claim 15 or 16, wherein the texture profiles each identify pixels of the portion having texture values within a particular range using a binary representation, wherein the binary representation uses binary values to distinguish, as a first of two binary values, the pixels of the portion having texture values within the particular range, from as a second of two binary values, the pixels of the image having texture values outside the particular range.
- 18. A method according to claim 17, when dependent on claim 16, comprising determining the position of the portion for the first texture profile by positioning the portion around a pixel having a binary value equal to the first binary value.
- 19. A method according to claim 17, when dependent on claim 16, comprising generating the one or more other texture profiles for portions positioned around a respective one or more pixels within the first texture profile having a binary value equal to the first binary value.
- 20. A method according to claim 18, when dependent on claim 16, comprising generating the one or more other texture profiles for portions positioned around a respective one or more pixels within the first texture profile which each (i) has a binary value equal to the first binary value and (ii) is connected, either directly or through one or more other pixels having the first binary value, to the pixel around which the first portion is positioned.
- 21. A method according to any of claims 16 to 20, wherein a set of predetermined particular ranges of texture values is defined and the method is performed for different particular ranges.
- 22. A method according to any of claims 16 to 20, wherein the particular range of texture values for the texture profile is defined as being a relative range of values around the texture value of the pixel around which the portion is positioned.
- 23. A method according to any of claims 16 to 22, wherein the method is performed for each pixel of the digital image.
- 24. A method according to any of claims 16 to 23, wherein the portion of the digital image for which a composite texture profile is generated is of a predetermined size which is the same size as the size of the composite texture profiles stored in the electronic library.
- 25. A method according to any of claims 14 to 24, wherein comparing the generated composite texture profile with one or more composite texture profiles in the electronic library to identify the best fitting composite texture profile uses a summed comparison value and comprises: for each possible combination of a pixel in the generated composite texture profile with a pixel in the composite texture profile from the library, calculating a comparison value by multiplying the two respective pixel values by each other and the distance between the two pixels; summing the comparison value for each pixel combination to determine a summed comparison value; and selecting the composite texture profile from the library with the lowest summed comparison value as the best fitting composite texture profile.
- 26. A method according to claim 25, further comprising subtracting the mutual minimum values of each profile from each pixel of each profile and using the resulting profiles for the calculation of the comparison values.
- 27. A method according to any preceding claim, wherein the digital image comprises medical data.
- 28. A method according to claim 27, wherein the digital image comprises medical data of soft tissue.
- 29. A system comprising a memory and a processor, wherein the processor is arranged to perform the method of any preceding claim.
- 30. A computer-readable medium having computer-executable instructions adapted to cause a computer system to perform the method of any of claims ito 28.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB0822715A GB2466069B (en) | 2008-12-12 | 2008-12-12 | Texture characterisation |
PCT/GB2009/002872 WO2010067084A1 (en) | 2008-12-12 | 2009-12-11 | Texture characterisation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB0822715A GB2466069B (en) | 2008-12-12 | 2008-12-12 | Texture characterisation |
Publications (3)
Publication Number | Publication Date |
---|---|
GB0822715D0 GB0822715D0 (en) | 2009-01-21 |
GB2466069A true GB2466069A (en) | 2010-06-16 |
GB2466069B GB2466069B (en) | 2010-11-03 |
Family
ID=40326030
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB0822715A Active GB2466069B (en) | 2008-12-12 | 2008-12-12 | Texture characterisation |
Country Status (2)
Country | Link |
---|---|
GB (1) | GB2466069B (en) |
WO (1) | WO2010067084A1 (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5224175A (en) * | 1987-12-07 | 1993-06-29 | Gdp Technologies, Inc. | Method for analyzing a body tissue ultrasound image |
EP1089232A2 (en) * | 1999-10-01 | 2001-04-04 | Samsung Electronics Co., Ltd. | Method for analyzing texture of digital image |
US6647132B1 (en) * | 1999-08-06 | 2003-11-11 | Cognex Technology And Investment Corporation | Methods and apparatuses for identifying regions of similar texture in an image |
-
2008
- 2008-12-12 GB GB0822715A patent/GB2466069B/en active Active
-
2009
- 2009-12-11 WO PCT/GB2009/002872 patent/WO2010067084A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5224175A (en) * | 1987-12-07 | 1993-06-29 | Gdp Technologies, Inc. | Method for analyzing a body tissue ultrasound image |
US6647132B1 (en) * | 1999-08-06 | 2003-11-11 | Cognex Technology And Investment Corporation | Methods and apparatuses for identifying regions of similar texture in an image |
EP1089232A2 (en) * | 1999-10-01 | 2001-04-04 | Samsung Electronics Co., Ltd. | Method for analyzing texture of digital image |
Also Published As
Publication number | Publication date |
---|---|
WO2010067084A1 (en) | 2010-06-17 |
GB0822715D0 (en) | 2009-01-21 |
GB2466069B (en) | 2010-11-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liang et al. | International benchmarking of terrestrial laser scanning approaches for forest inventories | |
Flood et al. | Using a U-net convolutional neural network to map woody vegetation extent from high resolution satellite imagery across Queensland, Australia | |
Zhu et al. | A flexible spatiotemporal method for fusing satellite images with different resolutions | |
US11521380B2 (en) | Shadow and cloud masking for remote sensing images in agriculture applications using a multilayer perceptron | |
Trygonis et al. | photoQuad: a dedicated seabed image processing software, and a comparative error analysis of four photoquadrat methods | |
Mairhofer et al. | RooTrak: automated recovery of three-dimensional plant root architecture in soil from X-ray microcomputed tomography images using visual tracking | |
US11250569B2 (en) | Systems and methods for functional imaging follow-up evaluation using deep neural network | |
Peeters et al. | Getis–Ord’s hot-and cold-spot statistics as a basis for multivariate spatial clustering of orchard tree data | |
Van Niekerk | A comparison of land unit delineation techniques for land evaluation in the Western Cape, South Africa | |
Pedroso et al. | A segmentation algorithm for the delineation of agricultural management zones | |
Kumar et al. | Automatic segmentation of liver and tumor for CAD of liver | |
US10311571B2 (en) | Image analysis method supporting illness development prediction for a neoplasm in a human or animal body | |
Packalen et al. | Predicting the spatial pattern of trees by airborne laser scanning | |
CN104094314B (en) | The automatic detection of row risk zones is mapped into using quantitative T1 | |
CN109377500B (en) | Image segmentation method based on neural network and terminal equipment | |
Ţălu | Texture analysis methods for the characterisation of biological and medical images | |
CN103562960B (en) | For generating the assigned unit between the image-region of image and element class | |
JP2008527546A (en) | Improved image segmentation method | |
Peng et al. | Object-based change detection from satellite imagery by segmentation optimization and multi-features fusion | |
US20140003686A1 (en) | Multimodality Image Segmentation of Volumetric Data Sets | |
CN104838422A (en) | Image processing device and method | |
Kallel et al. | An iterative possibilistic knowledge diffusion approach for blind medical image segmentation | |
KR102473298B1 (en) | Method for obtaining 3D model data of multiple components of an object | |
Wujanz et al. | Plane-based registration of several thousand laser scans on standard hardware | |
EP2863360B1 (en) | Multimodality image segmentation of volumetric data sets |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
732E | Amendments to the register in respect of changes of name or changes affecting rights (sect. 32/1977) |
Free format text: REGISTERED BETWEEN 20130801 AND 20130807 |