CN111105470A - Burn patient portrait segmentation method based on skin color detection - Google Patents
Burn patient portrait segmentation method based on skin color detection Download PDFInfo
- Publication number
- CN111105470A CN111105470A CN201911341331.0A CN201911341331A CN111105470A CN 111105470 A CN111105470 A CN 111105470A CN 201911341331 A CN201911341331 A CN 201911341331A CN 111105470 A CN111105470 A CN 111105470A
- Authority
- CN
- China
- Prior art keywords
- portrait
- skin color
- segmentation
- area
- color detection
- 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
Images
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
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
Abstract
The invention relates to a burn patient portrait segmentation method based on skin color detection. The method comprises the steps of adopting a depeplab model to carry out portrait segmentation on a burn patient, judging that the segmentation effect is poor if the area of the portrait occupying the whole image after the portrait segmentation is smaller than a threshold value, carrying out skin color detection on an original image by Gaussian filtering, removing small-area missing points appearing in the portrait by closed operation, then carrying out outline extraction by opencv, finding out large-area closed missing blocks, filling skin colors to obtain complete portrait and skin color-like parts, then adopting the depeplab model to carry out portrait segmentation on the complete portrait and skin color-like parts, removing non-portrait parts, and if the area of the obtained portrait segmentation area occupying the whole image is smaller than the threshold value, directly taking the complete portrait and skin color-like partial images after the skin color detection processing as the portrait segmentation result. The method and the device can accurately identify the portrait part from the complex environment.
Description
Technical Field
The invention belongs to the field of medical image segmentation, and particularly relates to a burn patient portrait segmentation method based on skin color detection.
Background
The existing portrait segmentation technology:
a portrait segmentation model formed by a portrait segmentation method and device (201910568826.0), a method for generating a portrait segmentation model, a video key frame extraction method (201910055748.4), a portrait segmentation method and device (201811333344.9) and the like only has a good segmentation effect on normal portraits and has a poor segmentation effect on burn patients. Compared with normal people, the skin color of burn patients changes, and the skin color is often interfered by wounds, gauze and the like, so that the image part of the burn patients is difficult to be accurately identified from the complex environment by the existing model.
Disclosure of Invention
The invention aims to provide a burn patient portrait segmentation method based on skin color detection, which can accurately identify a portrait part from a complex environment.
In order to achieve the purpose, the technical scheme of the invention is as follows: a burn patient portrait segmentation method based on skin color detection is characterized in that a depeplab model is adopted to segment a portrait of a burn patient, if the area of the portrait occupying the whole image after the portrait segmentation is found to be smaller than a threshold value, the segmentation effect is judged to be poor, Gaussian filtering is adopted to detect the skin color of an original image, a small-area missing point appearing in the portrait is removed through closed operation, then opencv is used for contour extraction, a large-area closed missing block is found out, skin color filling is carried out, a relatively complete portrait and a relatively skin color part are obtained, then the relatively complete portrait and the relatively skin color part are subjected to portrait segmentation through the depeplab model, a non-portrait part is removed, and a portrait segmentation result is obtained.
In an embodiment of the present invention, if a relatively complete portrait and a similar skin color part are obtained after skin color detection processing, and a depeplab model is used to perform portrait segmentation on the image after skin color processing to obtain that the area of the portrait segmentation area occupying the whole image is smaller than a threshold, the relatively complete portrait and the similar skin color part image after skin color detection processing are directly used as a portrait segmentation result.
In an embodiment of the present invention, the threshold is a percentage value of an artificially set segmentation area in an entire image area, and in this embodiment, the threshold is 10%.
Compared with the prior art, the invention has the following beneficial effects: the invention can accurately identify the portrait part from the image in the complex environment, and particularly aims at the portrait segmentation of the burn patient, separates the portrait and the similar skin color part by detecting and missing the skin color of the patient image, and improves the segmentation effect of the burn patient image.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
FIG. 2 is a graph showing the results of the experiment according to the present invention.
Detailed Description
The technical scheme of the invention is specifically explained below with reference to the accompanying drawings.
As shown in figure 1, the invention relates to a burn patient portrait segmentation method based on skin color detection, which firstly uses a deeplab model to segment the portrait of a burn patient. If the area of the portrait occupying the whole image after the portrait is segmented is too small, the segmentation effect is judged to be poor, Gaussian filtering is used for detecting the skin color of the original image, small-area missing points appearing in the portrait are removed through closed operation, then outline extraction is carried out through opencv, large-area closed missing blocks (mostly caused by bandages and wounds) are found out, skin color filling is carried out, a relatively complete portrait and a skin color-like part are obtained, then portrait segmentation is carried out on the part through deplab, a non-portrait part is removed, and if the area of the segmented image occupies the whole image to be too small, the image after the skin color detection processing is directly used as a portrait segmentation result.
As shown in fig. 2, the experimental result graph includes, from left to right, the original graph, the deplab segmentation graph, and the deplab segmentation graph after the skin color detection processing.
The above are preferred embodiments of the present invention, and all changes made according to the technical scheme of the present invention that produce functional effects do not exceed the scope of the technical scheme of the present invention belong to the protection scope of the present invention.
Claims (3)
1. A burn patient portrait segmentation method based on skin color detection is characterized in that a depeplab model is adopted to segment a portrait of a burn patient, if the area of the portrait occupying the whole image after portrait segmentation is found to be smaller than a threshold value, the segmentation effect is judged to be poor, Gaussian filtering is adopted to detect the skin color of an original image, a small-area missing point appearing in the portrait is removed through closed operation, then opencv is used for contour extraction, a large-area closed missing block is found out, skin color filling is carried out, a relatively complete portrait and a relatively skin color part are obtained, then portrait segmentation is carried out on the relatively complete portrait and the relatively skin color part through the depeplab model, a non-portrait part is removed, and a portrait segmentation result is obtained.
2. The method for segmenting the portrait of the burn patient based on the skin color detection as claimed in claim 1, wherein if the complete portrait and the skin-like color part are obtained after the skin color detection processing, and the portrait area of the image after the skin color processing is segmented by adopting a depeplab model to obtain that the area of the portrait accounts for less than a threshold value in the whole image, the complete portrait and the skin-like color part image after the skin color detection processing are directly used as the portrait segmentation result.
3. A method for skin color detection based segmentation of a human image of a burned patient according to claim 1 or 2, wherein the threshold is a percentage value of the manually set segmentation area to the whole image area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911341331.0A CN111105470B (en) | 2019-12-24 | 2019-12-24 | Burn patient portrait segmentation method based on skin color detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911341331.0A CN111105470B (en) | 2019-12-24 | 2019-12-24 | Burn patient portrait segmentation method based on skin color detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111105470A true CN111105470A (en) | 2020-05-05 |
| CN111105470B CN111105470B (en) | 2023-02-24 |
Family
ID=70423933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911341331.0A Active CN111105470B (en) | 2019-12-24 | 2019-12-24 | Burn patient portrait segmentation method based on skin color detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111105470B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102324036A (en) * | 2011-09-02 | 2012-01-18 | 北京新媒传信科技有限公司 | Obtain the method and apparatus of face complexion area in the image |
| CN105869159A (en) * | 2016-03-28 | 2016-08-17 | 联想(北京)有限公司 | Image segmentation method and apparatus |
| CN106845328A (en) * | 2016-11-09 | 2017-06-13 | 广州市贺氏办公设备有限公司 | A kind of Intelligent human-face recognition methods and system based on dual camera |
| CN106909883A (en) * | 2017-01-17 | 2017-06-30 | 北京航空航天大学 | A kind of modularization hand region detection method and device based on ROS |
| US20170337695A1 (en) * | 2015-01-30 | 2017-11-23 | Ventana Medical Systems, Inc. | Foreground segmentation and nucleus ranking for scoring dual ish images |
| CN109035257A (en) * | 2018-07-02 | 2018-12-18 | 百度在线网络技术（北京）有限公司 | portrait dividing method, device and equipment |
| CN110310285A (en) * | 2019-05-14 | 2019-10-08 | 武汉泓毅智云信息有限公司 | A kind of burn surface area calculation method accurately rebuild based on 3 D human body |
| CN110322435A (en) * | 2019-01-20 | 2019-10-11 | 北京工业大学 | A kind of gastric cancer pathological image cancerous region dividing method based on deep learning |
-
2019
- 2019-12-24 CN CN201911341331.0A patent/CN111105470B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102324036A (en) * | 2011-09-02 | 2012-01-18 | 北京新媒传信科技有限公司 | Obtain the method and apparatus of face complexion area in the image |
| US20170337695A1 (en) * | 2015-01-30 | 2017-11-23 | Ventana Medical Systems, Inc. | Foreground segmentation and nucleus ranking for scoring dual ish images |
| CN105869159A (en) * | 2016-03-28 | 2016-08-17 | 联想(北京)有限公司 | Image segmentation method and apparatus |
| CN106845328A (en) * | 2016-11-09 | 2017-06-13 | 广州市贺氏办公设备有限公司 | A kind of Intelligent human-face recognition methods and system based on dual camera |
| CN106909883A (en) * | 2017-01-17 | 2017-06-30 | 北京航空航天大学 | A kind of modularization hand region detection method and device based on ROS |
| CN109035257A (en) * | 2018-07-02 | 2018-12-18 | 百度在线网络技术（北京）有限公司 | portrait dividing method, device and equipment |
| CN110322435A (en) * | 2019-01-20 | 2019-10-11 | 北京工业大学 | A kind of gastric cancer pathological image cancerous region dividing method based on deep learning |
| CN110310285A (en) * | 2019-05-14 | 2019-10-08 | 武汉泓毅智云信息有限公司 | A kind of burn surface area calculation method accurately rebuild based on 3 D human body |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111105470B (en) | 2023-02-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107194919B (en) | Mobile phone screen defect detection method based on regular texture background reconstruction | |
| CN108109155A (en) | A kind of automatic threshold edge detection method based on improvement Canny | |
| CN106485696B (en) | A kind of detection method of the explosive dangerous material stitch defect based on machine vision | |
| CN104732536A (en) | Sub-pixel edge detection method based on improved morphology | |
| CN106446905A (en) | Surface crack texture extraction method based on fusion of seepage algorithm and adaptive Canny algorithm | |
| CN105092597B (en) | A kind of crack detecting method on hard plastic material surface | |
| CN105187721B (en) | A kind of the license camera and method of rapid extraction portrait feature | |
| CN106127193B (en) | A kind of facial image recognition method | |
| CN110276279B (en) | Method for detecting arbitrary-shape scene text based on image segmentation | |
| CN105447489B (en) | A kind of character of picture OCR identifying system and background adhesion noise cancellation method | |
| CN105701487A (en) | Vehicle logo positioning method | |
| CN110929695B (en) | Face recognition and pedestrian re-recognition association method | |
| CN104899905A (en) | Face image processing method and apparatus | |
| CN104766316A (en) | Novel lip segmentation algorithm for traditional Chinese medical inspection diagnosis | |
| CN113808027B (en) | Human body image processing method and device, electronic equipment and storage medium | |
| CN110348461A (en) | A kind of Surface Flaw feature extracting method | |
| CN104392211A (en) | Skin recognition method based on saliency detection | |
| CN107230200A (en) | A kind of method for extracting rotor coil contour feature | |
| CN111105470B (en) | Burn patient portrait segmentation method based on skin color detection | |
| CN111179156A (en) | Video beautifying method based on face detection | |
| Xu et al. | Improved Canny Edge Detection Operator | |
| CN108921010A (en) | A kind of pupil detection method and detection device | |
| CN109035272B (en) | Lung segmentation method and device based on CT image and computer readable storage medium | |
| JP2003256834A (en) | Face area extracting and face configuring element position judging device | |
| CN104778450B (en) | A kind of preprocess method and device of vehicle identification code image |
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 |