CN107220646B - Medical image character recognition enhancing method for removing background interference - Google Patents
Medical image character recognition enhancing method for removing background interference Download PDFInfo
- Publication number
- CN107220646B CN107220646B CN201710376310.7A CN201710376310A CN107220646B CN 107220646 B CN107220646 B CN 107220646B CN 201710376310 A CN201710376310 A CN 201710376310A CN 107220646 B CN107220646 B CN 107220646B
- Authority
- CN
- China
- Prior art keywords
- medical image
- medical
- character recognition
- background interference
- images
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 17
- 230000002708 enhancing effect Effects 0.000 title claims abstract description 7
- 239000003086 colorant Substances 0.000 abstract description 3
- 238000002595 magnetic resonance imaging Methods 0.000 description 5
- 238000010187 selection method Methods 0.000 description 2
- 210000001519 tissue Anatomy 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 210000004072 lung Anatomy 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 210000003625 skull Anatomy 0.000 description 1
- 238000003325 tomography Methods 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention relates to a method for recognizing and enhancing medical image characters without background interference, which comprises the following steps: firstly, selecting two adjacent and similar original medical images in the same sequence according to a preset rule; then, performing reverse phase processing on RGB colors of a medical image to generate a new medical image; subtracting the new medical image from another original medical image to obtain an absolute value; and finally, obtaining the medical image with enhanced character recognition through binarization processing. The method removes the background interference of the fault medical image through reverse phase superposition subtraction to obtain high-quality medical character information, thereby facilitating the classification of medical images and the identification of patients and doctors.
Description
Technical Field
The invention relates to the technical field of medical image recognition, in particular to a method for recognizing and enhancing characters of a medical image without background interference.
Background
Computed tomography (ct) is a continuous cross-sectional scan around a part of the human body with a highly sensitive detector using precisely collimated X-ray beams, gamma rays, ultrasound, etc. Magnetic Resonance imaging (mri), which is also a kind of tomographic imaging, obtains electromagnetic signals from a human body by using a magnetic Resonance phenomenon and reconstructs human body information.
In practice, most of CT and MRI can not store the text information in layers, the system marks the text information at four corners of a medical image, and parts of the text information are overlapped on the image, so that the accurate extraction of the text information is influenced, and the classification of the medical image is influenced. The textual information includes, but is not limited to, patient name, age, gender, exam number, visit number, time, camera, CT value, window width, FOV, KV, mAs, matrix, etc. In order to accurately extract the text information, it is necessary to reduce the interference of the image as much as possible.
Each examination contains one or more series of images, each series having tens or hundreds of medical images, based on the continuity of CT, MRI tomography, including the number of images and the layer thickness/distance set by the doctor at the time of taking the images. Wherein, the layer thickness refers to the thickness of the scanning layer, the layer distance refers to the distance between the centers of two layers, the layer thickness is generally set to be 2-5mm, and the layer distance is continuous, so the distance between two adjacent reconstructed images is 0-10 mm. Both CT and MRI use the change in density or water content of human tissue to form images, and the change in density or water content of human tissue is continuous. Therefore, the outlines and internal images of two consecutive scanning layers have greater similarity, and the thinner the layer thickness and the smaller the layer distance, the higher the similarity.
In view of the characteristics of CT and MRI tomographic images: the same sequence of picture labels are identical in the check number, visit number, name, age, gender, etc., and the absolute position and size of the individual images and color values of these information are fixed. The method of the invention removes the background interference of the fault medical image by reverse phase superposition subtraction to improve the recognition rate of the medical character information.
Disclosure of Invention
In order to solve the problems, the invention provides a method for identifying and enhancing the characters of the medical image by removing the background interference of the fault medical image, which greatly improves the identification of the character information of the medical image.
The invention realizes a medical image character recognition enhancing method for removing background interference by the following technical scheme, which comprises the following steps:
selecting two adjacent and similar original medical images in the same sequence according to a preset rule;
performing reverse phase processing on RGB color of a medical image to generate a new medical image;
subtracting the original medical image with the other original medical image to obtain an absolute value;
performing binarization processing to obtain a medical image with enhanced character recognition;
the inversion processing refers to inverting the color hue of the medical image.
Further, the preset rule is that two similar medical images are selected according to the characteristics of the shooting part.
The invention removes the background interference of the fault medical image by reverse phase superposition subtraction to obtain high-quality medical character information, thereby facilitating the classification of medical images and the identification of patients and doctors.
Drawings
FIG. 1: a craniocerebral CT image of the first embodiment.
FIG. 2: CT image of mediastinal window of example two.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
The method of the present invention is described in detail with reference to a craniocerebral CT image. The method comprises the following steps:
(1) selecting two adjacent medical images of the same sequence of the patient craniocerebral CT examination, wherein the selection method comprises the following steps: based on the fact that the skull is an oval sphere, the sectional areas of the initial part and the ending part of the same sequence of the sphere are changed greatly, and the method provided by the invention has the advantages that the more similar the two pictures are, the better the effect is, and therefore, the two continuous slice images in the middle part are selected to be better. Referring to fig. 1, a partial slice of a brain CT sequence containing 25 images was selected. Where A and B are two consecutive pieces of the initial portion, C and D are two consecutive pieces of the middle portion, and E and F are two consecutive pieces of the ending portion. By comparison, it can be seen that the similarities of a and B at the beginning and E and F at the end are low, while the similarities of C and D at the middle are high. Therefore, the method is suitable for selecting C and D as processing objects.
(2) And performing reverse phase processing on RGB colors of a medical image to generate a new medical image.
(3) And (3) subtracting the new medical image from the other original medical image in the step (1) according to the RGB values of the pixels with the same coordinates, and taking the absolute value.
(4) And (4) carrying out binarization processing on the image processed in the step (3), setting the threshold value to be 255 and setting the threshold value to be 0 when the threshold value is smaller than 255, and obtaining the medical image with enhanced character recognition.
The method of the present invention is described in detail with reference to the second longitudinal separation window CT image of the embodiment. The method comprises the following steps:
(1) two adjacent medical images of the same sequence of the patient longitudinal compartment window CT examination are selected, and the selection method comprises the following steps: the image based on the mediastinal window has a large change at the beginning and a small change near the end, so that it is better to select two consecutive slices near the end. Referring to fig. 2, a partial slice of a lung CT mediastinal window sequence containing 65 images was selected. Where A and B are two consecutive pieces of the initial portion, C and D are two consecutive pieces of the middle portion, and E and F are two consecutive pieces of the ending portion. By comparison, it can be seen that the similarities of a and B in the beginning portion and C and D in the middle portion are low, and the similarities of E and F in the end portion are high. Therefore, the method is suitable for selecting E and F as processing objects.
(2) And performing reverse phase processing on RGB colors of a medical image to generate a new medical image.
(3) And (3) subtracting the new medical image from the other original medical image in the step (1) according to the RGB values of the pixels with the same coordinates, and taking the absolute value.
(4) And (4) carrying out binarization processing on the image processed in the step (3), setting the threshold value to be 255 and setting the threshold value to be 0 when the threshold value is smaller than 255, and obtaining the medical image with enhanced character recognition.
Although the present invention has been described with reference to the preferred embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but is intended to cover various modifications, equivalents and alternatives falling within the spirit and scope of the invention.
Claims (2)
1. A medical image character recognition enhancing method for removing background interference is characterized by comprising the following steps:
selecting two adjacent and similar original medical images in the same sequence according to a preset rule;
performing reverse phase processing on RGB color of a medical image to generate a new medical image;
subtracting the original medical image with the other original medical image to obtain an absolute value;
performing binarization processing to obtain a medical image with enhanced character recognition;
the inversion processing refers to inverting the color hue of the medical image.
2. The method according to claim 1, wherein the predetermined rule is to select two similar medical images according to the characteristics of the photographed part.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710376310.7A CN107220646B (en) | 2017-05-25 | 2017-05-25 | Medical image character recognition enhancing method for removing background interference |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710376310.7A CN107220646B (en) | 2017-05-25 | 2017-05-25 | Medical image character recognition enhancing method for removing background interference |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107220646A CN107220646A (en) | 2017-09-29 |
| CN107220646B true CN107220646B (en) | 2020-04-14 |
Family
ID=59944541
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710376310.7A Active CN107220646B (en) | 2017-05-25 | 2017-05-25 | Medical image character recognition enhancing method for removing background interference |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107220646B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109785940B (en) * | 2018-12-28 | 2023-02-03 | 苏州承泽医疗科技有限公司 | Method for typesetting medical image film |
| CN110060022A (en) * | 2019-03-12 | 2019-07-26 | 杭州华卓信息科技有限公司 | A kind of medicine film Intelligent printing method and system |
| CN110909674B (en) * | 2019-11-21 | 2024-01-05 | 清华大学苏州汽车研究院（吴江） | Traffic sign recognition method, device, equipment and storage medium |
| CN111028186B (en) * | 2019-11-25 | 2023-07-04 | 泰康保险集团股份有限公司 | Image enhancement method and device |
| CN112348023A (en) * | 2020-10-28 | 2021-02-09 | 南阳柯丽尔科技有限公司 | Background and character separation method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003058826A (en) * | 2001-08-10 | 2003-02-28 | Konica Corp | Character information extraction device and its method, program and storage medium |
| CN1811794A (en) * | 2006-03-03 | 2006-08-02 | 清华大学 | Greyscale image partition method based on light distribution character of reflection or penetrance |
| CN102982238A (en) * | 2012-11-13 | 2013-03-20 | 深圳市环球博洋机电科技有限公司 | Three-dimensional kidney neoplasm surgery simulation method and platform based on computed tomography (CT) film |
| CN103337175A (en) * | 2013-06-22 | 2013-10-02 | 太仓博天网络科技有限公司 | Vehicle type recognition system based on real-time video steam |
| CN106683085A (en) * | 2016-12-23 | 2017-05-17 | 浙江大学 | CT image spine and spinal dura mater automation detection method |
-
2017
- 2017-05-25 CN CN201710376310.7A patent/CN107220646B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003058826A (en) * | 2001-08-10 | 2003-02-28 | Konica Corp | Character information extraction device and its method, program and storage medium |
| CN1811794A (en) * | 2006-03-03 | 2006-08-02 | 清华大学 | Greyscale image partition method based on light distribution character of reflection or penetrance |
| CN102982238A (en) * | 2012-11-13 | 2013-03-20 | 深圳市环球博洋机电科技有限公司 | Three-dimensional kidney neoplasm surgery simulation method and platform based on computed tomography (CT) film |
| CN103337175A (en) * | 2013-06-22 | 2013-10-02 | 太仓博天网络科技有限公司 | Vehicle type recognition system based on real-time video steam |
| CN106683085A (en) * | 2016-12-23 | 2017-05-17 | 浙江大学 | CT image spine and spinal dura mater automation detection method |
Non-Patent Citations (2)
| Title |
|---|
| CT 图像压缩在远程医疗传输系统中的应用研究;涂咏涛;《自动化与仪器仪表》;20170331(第3期);179-184 * |
| 基于塔型结构和ＬＩＰ模型的图像增强方法;田小平 等;《西安邮电大学学报》;20150531;第20卷(第3期);48-52 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107220646A (en) | 2017-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107220646B (en) | Medical image character recognition enhancing method for removing background interference | |
| US11062449B2 (en) | Method and system for extracting vasculature | |
| Leynes et al. | Direct pseudoCT generation for pelvis PET/MRI attenuation correction using deep convolutional neural networks with multi-parametric MRI: zero echo-time and Dixon deep pseudoCT (ZeDD-CT) | |
| CN106133790B (en) | Method and device for generating one or more computed tomography images based on magnetic resonance images with the aid of tissue type separation | |
| US8290568B2 (en) | Method for determining a property map of an object, particularly of a living being, based on at least a first image, particularly a magnetic resonance image | |
| US20130190602A1 (en) | 2d3d registration for mr-x ray fusion utilizing one acquisition of mr data | |
| US9675317B2 (en) | Interface identification apparatus and method | |
| US9336613B2 (en) | Apparatus for generating assignments between image regions of an image and element classes | |
| US20160292844A1 (en) | Medical image registration apparatus, medical image registration method, and medical image registration program | |
| CN111602173A (en) | Tomographic data analysis | |
| CN108806776A (en) | A method of the Multimodal medical image based on deep learning | |
| EP3107031A1 (en) | Method, apparatus and system for spine labeling | |
| US8644608B2 (en) | Bone imagery segmentation method and apparatus | |
| Ganasala et al. | Medical image fusion based on Frei-Chen masks in NSST domain | |
| Abdallah | Increasing of edges recognition in cardiac scintography for ischemic patients | |
| US10307124B2 (en) | Image display device, method, and program for determining common regions in images | |
| KR102136107B1 (en) | Apparatus and method for alignment of bone suppressed chest x-ray image | |
| EP4102461A2 (en) | Medical image registration method and apparatus | |
| Pal et al. | Panoptic Segmentation and Labelling of Lumbar Spine Vertebrae using Modified Attention Unet | |
| US11776154B2 (en) | Method and device for medical imaging for representing a 3D volume containing at least one introduced foreign object | |
| Longuefosse et al. | Lung CT Synthesis Using GANs with Conditional Normalization on Registered Ultrashort Echo-Time MRI | |
| EP4336452A1 (en) | Computer-implemented method for processing spectral computed tomography (spectral ct) data, computer program and spectral ct system | |
| Ruby Grace et al. | A Study On Image Fusion Techniques Of Complementary Medical Images. | |
| Fouquier et al. | Subcutaneous adipose tissue segmentation in whole-body mri of children | |
| Kriston et al. | Segmentation of multiple organs in Computed Tomography and Magnetic Resonance Imaging measurements |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder | ||
| CP02 | Change in the address of a patent holder |
Address after: 311200 floor 4-6, East Building e, Hangzhou Bay Information Port, Xiaoshan District, Hangzhou City, Zhejiang Province Patentee after: HANGZHOU JIANPEI TECHNOLOGY Co.,Ltd. Address before: 310018 No. 3, No. 3, Hangzhou economic and Technological Development Zone, Hangzhou, Zhejiang Patentee before: HANGZHOU JIANPEI TECHNOLOGY Co.,Ltd. |