GB2589478A - Segmenting irregular shapes in images using deep region growing - Google Patents
Segmenting irregular shapes in images using deep region growing Download PDFInfo
- Publication number
- GB2589478A GB2589478A GB2019774.5A GB202019774A GB2589478A GB 2589478 A GB2589478 A GB 2589478A GB 202019774 A GB202019774 A GB 202019774A GB 2589478 A GB2589478 A GB 2589478A
- Authority
- GB
- United Kingdom
- Prior art keywords
- node
- image
- nodes
- layer
- value
- 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
- 230000001788 irregular Effects 0.000 title 1
- 238000013528 artificial neural network Methods 0.000 claims abstract 19
- 238000000034 method Methods 0.000 claims 27
- 230000006870 function Effects 0.000 claims 8
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
Abstract
A system for determining a region of interest in an image. The system includes a memory and an electronic processor. The electronic processor included in the system is connected to the memory and is configured to initialize internal states of nodes of a spatial lattice. Each node of the spatial lattice corresponds to a pixel of the image and is connected to at least one node representing a neighboring pixel of the image. The electronic processor is also configured to iteratively update, using a neural network, the internal states of each nodes in the spatial lattice using spatially gated propagation and identify the region of interest within the image based on the internal states of the nodes at a convergence of the spatial lattice. The electronic processor is configured to creating an image pyramid for the image.
Claims (46)
1. A method for identifying an object of interest in a medical image, the method comprising: initializing internal states of nodes of a spatial lattice, wherein each node corresponds to a pixel of the medical image and is connected to at least one node representing a neighboring pixel of the medical image; iteratively updating, using a neural network, the internal states of the nodes in the spatial lattice using spatially gated propagation, wherein at each iteration each node updates its internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, and a new value of the node; and identifying the object of interest within the medical image based on the values of the nodes at a convergence of the spatial lattice.
2. The method according to claim 1, wherein iteratively updating, using a neural network, the internal states of the nodes includes updating a value in a vector of values associated with the internal states of the nodes.
3. The method according to claim 2, wherein the values in the vector of values include a value representing the brightness of the pixel corresponding to the node and a value representing the internal state of the node.
4. The method according to claim 1 , wherein a convolution involving previous internal states of the nodes is performed for each iteration.
5. The method according to claim 1, wherein the method further includes performing, in a first iteration, convolutions on each value representing a brightness of each pixel.
6. The method according to claim 1, wherein identifying an object of interest within the medical image based on the values of the nodes at a convergence of the spatial lattice includes using a final layer of the neural network to calculate a probability that each pixel is included in the object of interest based on a value included in a vector of values associated with each pixel; and determining, for each pixel, if the calculated probability is above a predetermined threshold.
7. The method according to claim 1, wherein each node updates its internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, and a new value of the node using a squashing function.
8. The method according to claim 1, wherein the neighboring node is one selected from a group consisting of a node that represents a pixel that is directly above, directly below, to the right of, and to the left of a pixel represented by the node.
9. The method according to claim 1, the method further comprising generating an image pyramid with a plurality of layers, wherein each successive layer represents the medical image with fewer values.
10 The method according to claim 9, the method further comprising concatenating values from a plurality of layers of the image pyramid in each iteration.
11. A system for determining a region of interest in an image, the system comprising a memory; and an electronic processor, connected to the memory and configured to initialize internal states of nodes of a spatial lattice, wherein each node corresponds to a pixel of the image and is connected to at least one node representing a neighboring pixel of the image, iteratively update, using a neural network, the internal states of each nodes in the spatial lattice using spatially gated propagation; and identify the region of interest within the image based on the internal states of the nodes at a convergence of the spatial lattice.
12. The system according to claim 11, wherein the electronic processor is configured to update the internal states of the nodes by, at each iteration, updating the internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, or a new value of the node.
13. The system according to claim 11, wherein the electronic processor is configured to iteratively update, using a neural network, the internal states of the nodes by updating a value in a vector of values associated with the internal states of the nodes.
14. The system according to claim 13, wherein the values in the vector of values include a value representing the brightness of a pixel corresponding to the node and a value representing the internal state of the node.
15. The system according to claim 11, wherein the electronic processor is further configured to perform, in each iteration, a convolution involving previous internal states of the nodes.
16. The system according to claim 11, wherein the electronic processor is further configured to perform, in the first iteration, convolutions on each value representing a brightness of each pixel.
17. The system according to claim 11, wherein the electronic processor is configured to identify an object of interest within the image based on the values of the nodes at a convergence of the spatial lattice by using a final layer of the neural network to calculate a probability that each pixel is included in the object of interest based on the vector associated with each pixel, and determining, for each pixel, if the calculated probability is above a predetermined threshold.
18. The system according to claim 12, wherein the electronic processor is configured to update the internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, or a new value of the node by using a squashing function.
19. The system according to claim 12, wherein the neighboring node is one selected from a group consisting of a node that represents a pixel that is directly above, directly below, to the right of, and to the left of the pixel represented by the node.
20. Non-transitory computer-readable medium storing instructions that, when executed with an electronic processor, perform a set of functions, the set of functions comprising: initializing internal states of nodes of a spatial lattice, wherein each node represents a pixel of an image and is connected to at least one neighboring pixel of the image; iteratively updating, using a neural network, the internal states of the nodes in the spatial lattice using spatially gated propagation, wherein at each iteration each node updates its internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, or a new value of the node; and identifying an object of interest within the image based on the values of the nodes at a convergence of the spatial lattice.
21. The non-transitory computer-readable medium according to claim 20, wherein iteratively updating, using a neural network, the internal states of the nodes includes updating a value in a vector of values associated with the internal states of the nodes.
22. The non-transitory computer-readable medium according to claim 20, wherein identifying an object of interest within the image based on the values of the nodes at a convergence of the spatial lattice includes using a final layer in the neural network to calculate a probability that each pixel is included in the object of interest based on the vector associated with each pixel; and determining, for each pixel, if the calculated probability is above a predetermined threshold.
23. A method for identifying an object of interest in a medical image, the method comprising: creating an image pyramid for the medical image, wherein the image pyramid includes a plurality of layers, each layer includes a plurality of values, each value represents a block of one or more pixels in the medical image, and each successive layer includes fewer values than a most previous layer; for each layer of the image pyramid; initializing internal states of nodes of a spatial lattice, wherein each node in the spatial lattice represents a block of one or more pixels in the medical image and is connected to at least one node representing a neighboring block of one or more pixels in the medical image; and iteratively updating, using a neural network, the internal states of the nodes in the spatial lattice using spatially gated propagation, wherein at each iteration each node updates its internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, and a new value of the node; and identifying the object of interest within the medical image based on the values of the nodes at a convergence of the spatial lattice having nodes representing the values included in a first layer of the image pyramid.
24. The method according to claim 23, wherein iteratively updating, using a neural network, the internal states of the nodes includes updating a value in a vector of values associated with the internal states of the nodes.
25. The method according to claim 23, the method further comprising performing, at each iteration for each layer of the image pyramid, a first convolution involving a first concatenation of previous internal states of the nodes representing the values included in a layer of the image pyramid and the values included in the layer of the image pyramid, and storing results of performing the first convolution.
26. The method according to claim 25, the method further comprising performing, at each iteration for each layer of the image pyramid, a second convolution involving a second concatenation of results of performing the first convolution for a current layer of the image pyramid, a layer of the image pyramid directly above the current layer of the image pyramid, and a layer of the image pyramid directly below the current layer of the image pyramid.
27. The method according to claim 23, wherein creating the image pyramid includes performing convolutions on each value representing a brightness of each block of one or more pixels in the medical image, wherein each convolution involving a reduction of dimensions of input medical image data produces values that are used to represent the medical image in a next layer of the image pyramid.
28. The method according to claim 23, wherein each value representing the medical image in the first layer of the image pyramid corresponds to a pixel in the medical image.
29. The method according to claim 28, wherein identifying the object of interest within the medical image based on the values of the nodes at a convergence of the spatial lattice having nodes representing the values included in a first layer of the image pyramid includes using a final layer of the neural network to calculate a probability that each pixel in the medical image is included in the object of interest based on a value included in each vector of values associated with a node representing the values included in a first layer of the image pyramid; and determining, for each pixel, if the calculated probability is above a predetermined threshold.
30. The method according to claim 26, wherein each node updates its internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, and a new value of the node includes using a squashing function and results of performing the second convolution.
31. The method according to claim 23, wherein the neighboring node is one selected from a group consisting of a node that represents a block of one or more pixels that is directly above, directly below, to the right of, and to the left of a block of one or more pixels represented by the node.
32. The method according to claim 23, wherein representing the medical image with fewer values creates a medical image with a lower resolution.
33. A system for determining a region of interest in an image, the system comprising a memory; and an electronic processor, connected to the memory and configured to: create an image pyramid for the image, the image pyramid including a plurality of layers, for each layer of the image pyramid, initialize internal states of nodes of a spatial lattice, wherein each node represents a block of one or more pixels in the image and is connected to at least one node representing a neighboring block of one or more pixels in the image, and iteratively update, using a neural network, the internal states of the nodes in the spatial lattice using spatially gated propagation; and identify the region of interest within the image based on the internal states of the nodes at a convergence of the spatial lattice having nodes representing values included in a first layer of the image pyramid.
34. The system according to claim 33, wherein each successive layer of the plurality of layers included in the image pyramid represents the image at a lower resolution than an image represented in a most previous layer of the image pyramid.
35. The system according to claim 34, wherein the electronic processor is configured to represent the image at a lower resolution by representing the image with fewer values.
36. The system according to claim 33, wherein the electronic processor is configured to update the internal states of the nodes by, at each iteration, deciding for each node whether to maintain a value of the node from a previous iteration, to set a value of the node to a value of a neighboring node from a previous iteration, or set a new value of the node.
37. The system according to claim 33, wherein the electronic processor is configured to iteratively update, using a neural network, the internal states of the nodes by updating a value in a vector of values associated with the internal states of the nodes.
38. The system according to claim 35, wherein the electronic processor is configured to perform, at each iteration for each layer of the image pyramid, a first convolution involving a first concatenation of previous internal states of the nodes representing the values included in the layer of the image pyramid and the values included in the layer of the image pyramid, and store results of performing the first convolution.
39. The system according to claim 38, wherein the electronic processor is configured to perform, at each iteration for each layer of the image pyramid, a second convolution involving a second concatenation of results of performing the first convolution for a current layer of the image pyramid, a layer of the image pyramid directly above the current layer of the image pyramid, and a layer of the image pyramid directly below the current layer of the image pyramid.
40. The system according to claim 34, wherein the electronic processor is further configured to perform, in the first iteration, convolutions on each value representing a brightness of each block of one or more pixels in the image, wherein each convolution involving a reduction of dimensions of input image data produces values that are used to represent the image in a next layer of the image pyramid.
41. The system according to claim 33, wherein the electronic processor is configured to identify an object of interest within the image based on the values of the nodes at a convergence of the spatial lattice having nodes representing values included in a first layer of the image pyramid by using a final layer of the neural network to calculate a probability that each pixel in the image is included in the object of interest based on each vector associated with a node representing the values included in a first layer of the image pyramid, and determining, for each pixel, if the calculated probability is above a predetermined threshold.
42. The system according to claim 39, wherein the electronic processor is configured to update the internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, or a new value of the node by using a squashing function and results of performing the second convolution.
43. The system according to claim 36, wherein the neighboring node is one selected from the group consisting of a node that represents a block of one or more pixels in the image that is directly above, directly below, to the right of, and to the left of the block of one or more pixels in the image represented by the node.
44. Non-transitory computer-readable medium storing instructions that, when executed with an electronic processor, perform a set of functions, the set of functions comprising: creating an image pyramid for an image, wherein the image pyramid includes a plurality of layers, each layer includes a plurality of values, each value represents a block of one or more pixels in the image, and each successive layer includes fewer values than a most previous layer; for each layer of the image pyramid; initializing internal states of nodes of a spatial lattice, wherein each node represents a block of one or more pixels in the image and is connected to at least one node representing a neighboring block of one or more pixels in of the image; and iteratively updating, using a neural network, the internal states of the nodes in the spatial lattice using spatially gated propagation, wherein at each iteration each node updates its internal state based on at least one selected from the group consisting of a value of the node from a previous iteration, a value of a neighboring node from the previous iteration, or a new value of the node; and identifying an object of interest within the image based on the values of the nodes at a convergence of the spatial lattice having nodes representing the values included in a first layer of the image pyramid.
45. The non-transitory computer-readable medium according to claim 44, wherein iteratively updating, using a neural network, the internal states of the nodes includes updating a value in a vector of values associated with the internal states of the nodes.
46. The non-transitory computer-readable medium according to claim 44, wherein identifying an object of interest within the image based on the values of the nodes at a convergence of the spatial lattice having nodes representing the values included in a first layer of the image pyramid includes using a final layer in the neural network to calculate a probability that each pixel in the image is included in the object of interest based on the vector associated with a node representing the values included in a first layer of the image pyramid; and determining, for each pixel, if the calculated probability is above a predetermined threshold.
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/014,801 US10643092B2 (en) | 2018-06-21 | 2018-06-21 | Segmenting irregular shapes in images using deep region growing with an image pyramid |
US16/014,785 US10776923B2 (en) | 2018-06-21 | 2018-06-21 | Segmenting irregular shapes in images using deep region growing |
PCT/IB2019/053923 WO2019243910A1 (en) | 2018-06-21 | 2019-05-13 | Segmenting irregular shapes in images using deep region growing |
Publications (3)
Publication Number | Publication Date |
---|---|
GB202019774D0 GB202019774D0 (en) | 2021-01-27 |
GB2589478A true GB2589478A (en) | 2021-06-02 |
GB2589478B GB2589478B (en) | 2022-05-25 |
Family
ID=68983773
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
GB2019774.5A Active GB2589478B (en) | 2018-06-21 | 2019-05-13 | Segmenting irregular shapes in images using deep region growing |
Country Status (5)
Country | Link |
---|---|
JP (1) | JP2021527859A (en) |
CN (1) | CN112189217A (en) |
DE (1) | DE112019001959T5 (en) |
GB (1) | GB2589478B (en) |
WO (1) | WO2019243910A1 (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116546340B (en) * | 2023-07-05 | 2023-09-19 | 华中师范大学 | High-speed CMOS pixel detector |
CN116894841B (en) * | 2023-09-08 | 2023-11-28 | 山东天鼎舟工业科技有限公司 | Visual detection method for quality of alloy shell of gearbox |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170372475A1 (en) * | 2016-06-23 | 2017-12-28 | Siemens Healthcare Gmbh | Method and System for Vascular Disease Detection Using Recurrent Neural Networks |
US20180033144A1 (en) * | 2016-09-21 | 2018-02-01 | Realize, Inc. | Anomaly detection in volumetric images |
CN107832807A (en) * | 2017-12-07 | 2018-03-23 | 深圳联影医疗科技有限公司 | A kind of image processing method and system |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103236058B (en) * | 2013-04-25 | 2016-04-13 | 内蒙古科技大学 | Obtain the method for volume of interest of four-dimensional heart image |
US9972093B2 (en) * | 2015-03-30 | 2018-05-15 | Siemens Healthcare Gmbh | Automated region of interest detection using machine learning and extended Hough transform |
US10332509B2 (en) * | 2015-11-25 | 2019-06-25 | Baidu USA, LLC | End-to-end speech recognition |
US10402700B2 (en) * | 2016-01-25 | 2019-09-03 | Deepmind Technologies Limited | Generating images using neural networks |
-
2019
- 2019-05-13 JP JP2020556276A patent/JP2021527859A/en active Pending
- 2019-05-13 DE DE112019001959.7T patent/DE112019001959T5/en active Pending
- 2019-05-13 CN CN201980033048.3A patent/CN112189217A/en active Pending
- 2019-05-13 GB GB2019774.5A patent/GB2589478B/en active Active
- 2019-05-13 WO PCT/IB2019/053923 patent/WO2019243910A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170372475A1 (en) * | 2016-06-23 | 2017-12-28 | Siemens Healthcare Gmbh | Method and System for Vascular Disease Detection Using Recurrent Neural Networks |
US20180033144A1 (en) * | 2016-09-21 | 2018-02-01 | Realize, Inc. | Anomaly detection in volumetric images |
CN107832807A (en) * | 2017-12-07 | 2018-03-23 | 深圳联影医疗科技有限公司 | A kind of image processing method and system |
Also Published As
Publication number | Publication date |
---|---|
WO2019243910A1 (en) | 2019-12-26 |
GB2589478B (en) | 2022-05-25 |
DE112019001959T5 (en) | 2021-01-21 |
GB202019774D0 (en) | 2021-01-27 |
CN112189217A (en) | 2021-01-05 |
JP2021527859A (en) | 2021-10-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107766794B (en) | Image semantic segmentation method with learnable feature fusion coefficient | |
US20200151858A1 (en) | Image contrast enhancement method and device, and storage medium | |
JP5045371B2 (en) | Foreground / background classification apparatus, method, and program for each pixel of moving image | |
CN108171663B (en) | Image filling system of convolutional neural network based on feature map nearest neighbor replacement | |
JP6463101B2 (en) | Region dividing apparatus and method | |
CN108629291B (en) | Anti-grid effect human face depth prediction method | |
CN109829868B (en) | Lightweight deep learning model image defogging method, electronic equipment and medium | |
CN111179196B (en) | Multi-resolution depth network image highlight removing method based on divide-and-conquer | |
CN104851133B (en) | A kind of image adaptive mess generation variational method | |
GB2589478A (en) | Segmenting irregular shapes in images using deep region growing | |
CN109345525B (en) | Ghost-removed high-dynamic-range image quality evaluation method | |
CN110033009A (en) | The method of image data is handled in connection network | |
CN113191983A (en) | Image denoising method and device based on deep learning attention mechanism | |
KR101563569B1 (en) | Learnable Dynamic Visual Image Pattern Recognition System and Method | |
CN111553296A (en) | Two-value neural network stereo vision matching method based on FPGA | |
CN111612721B (en) | Image restoration model training method and device and satellite image restoration method and device | |
JP2020014042A (en) | Image quality evaluation device, learning device and program | |
CN113256519A (en) | Image restoration method, apparatus, storage medium, and program product | |
CN105096309B (en) | A kind of edge detection method and device based on X-ray | |
CN111539517A (en) | Graph convolution neural network generation method based on graph structure matrix characteristic vector | |
CN109791687B (en) | Image restoration on arbitrary curved surfaces | |
CN113222867B (en) | Image data enhancement method and system based on multi-template image | |
US20220366539A1 (en) | Image processing method and apparatus based on machine learning | |
KR102201211B1 (en) | Method for analysis of blind ends in tunnels using artificial intelligence and extraction of joints | |
US11393069B2 (en) | Image processing apparatus, image processing method, and computer readable recording medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
746 | Register noted 'licences of right' (sect. 46/1977) |
Effective date: 20220613 |
|
732E | Amendments to the register in respect of changes of name or changes affecting rights (sect. 32/1977) |
Free format text: REGISTERED BETWEEN 20230615 AND 20230621 |
|
732E | Amendments to the register in respect of changes of name or changes affecting rights (sect. 32/1977) |
Free format text: REGISTERED BETWEEN 20231123 AND 20231129 |