CN112102235B - Human body part recognition method, computer device, and storage medium - Google Patents
Human body part recognition method, computer device, and storage medium Download PDFInfo
- Publication number
- CN112102235B CN112102235B CN202010787578.1A CN202010787578A CN112102235B CN 112102235 B CN112102235 B CN 112102235B CN 202010787578 A CN202010787578 A CN 202010787578A CN 112102235 B CN112102235 B CN 112102235B
- Authority
- CN
- China
- Prior art keywords
- human body
- image sequence
- body part
- target image
- label corresponding
- 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 88
- 210000000056 organ Anatomy 0.000 claims description 26
- 238000004590 computer program Methods 0.000 claims description 15
- 238000013507 mapping Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 8
- 239000010410 layer Substances 0.000 description 124
- 239000000284 extract Substances 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 10
- 230000000694 effects Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 210000001015 abdomen Anatomy 0.000 description 4
- 210000003484 anatomy Anatomy 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000004195 computer-aided diagnosis Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000002591 computed tomography Methods 0.000 description 3
- 238000002595 magnetic resonance imaging Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011478 gradient descent method Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 239000002356 single layer Substances 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000013170 computed tomography imaging Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening 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
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/38—Registration of image sequences
-
- 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
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10104—Positron emission tomography [PET]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Radiology & Medical Imaging (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The application relates to a human body part identification method, computer equipment and storage medium. The method comprises the following steps: acquiring an image sequence to be identified, and extracting a preset number of image layers from the image sequence to be identified; inputting each image layer into a preset human body identification model to obtain a human body part label corresponding to each image layer; fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified; obtaining a human body part label corresponding to the target image sequence according to the human body part recognition relationship, or obtaining a human body position label corresponding to the target image sequence according to the human body part recognition relationship; the target image sequence is any sequence in the image sequence to be identified. The accuracy of obtaining the human body part label or the human body position label corresponding to the target image sequence is improved.
Description
Technical Field
The present application relates to the field of medical images, and in particular, to a human body part recognition method, a computer device, and a storage medium.
Background
With the development of computer-aided diagnosis technology, the human body part covered by the medical image needs to be identified before the computer-aided diagnosis is performed, so that the human body part identification is particularly important in the computer-aided diagnosis.
In the traditional technology, the characteristics of the medical image to be analyzed are extracted, and the human body part covered by the medical image to be analyzed is identified through a classifier; or dividing the human body part into a preset number of areas according to the specific anatomical structure of the human body part, numbering the areas, and identifying the human body part covered by the medical image to be analyzed by adopting a neural network method.
However, the conventional human body part recognition method has a problem of low recognition accuracy.
Disclosure of Invention
In view of the foregoing, it is desirable to provide a human body part recognition method, a computer device, and a storage medium capable of improving accuracy of human body part recognition.
A method of human body part identification, the method comprising:
acquiring an image sequence to be identified, and extracting a preset number of image layers from the image sequence to be identified;
inputting each image layer into a preset human body identification model to obtain a human body part label corresponding to each image layer;
fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified;
obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
In one embodiment, the obtaining the human body part label corresponding to the target image sequence according to the human body part identification relationship and the human body position label corresponding to the target image sequence includes:
acquiring a human body position label corresponding to the target image sequence;
and predicting the human body part label corresponding to the target image sequence according to the human body position label corresponding to the target image sequence and the human body part identification relationship to obtain the human body part label of the target image sequence.
In one embodiment, the acquiring the human body position tag corresponding to the target image sequence includes:
and obtaining a human body position label corresponding to the target image sequence according to the position information of the target image sequence in the image sequence to be identified.
In one embodiment, the obtaining the human body position tag corresponding to the target image sequence according to the human body position recognition relationship and the human body position tag corresponding to the target image sequence includes:
acquiring a human body part label corresponding to the target image sequence; the label of the human body part corresponding to the target image sequence is the label of the human body part to which the target image sequence belongs;
and predicting the position label corresponding to the target image sequence according to the human body part label corresponding to the target image sequence and the human body part identification relationship to obtain the position label corresponding to the target image sequence.
In one embodiment, the acquiring the human body part tag corresponding to the target image sequence includes:
and obtaining a human body part label corresponding to the target image sequence according to the human body part label sequence to which the target image sequence belongs.
In one embodiment, the fitting method comprises a linear fitting method or a nonlinear fitting method.
In one embodiment, the extracting a preset number of image layers from the image sequence to be identified includes:
randomly extracting the preset number of image layers from the image sequence to be identified, or equally extracting the preset number of image layers from the image sequence to be identified according to a preset extraction interval.
In one embodiment, the method further comprises:
and obtaining the human organ to which the target image sequence belongs according to the mapping relation between the preset human organ and the human body part label or the mapping relation between the preset human organ and the human body position label.
A body part identification device, the device comprising:
the first acquisition module is used for acquiring an image sequence to be identified and extracting a preset number of image layers from the image sequence to be identified;
the second acquisition module is used for inputting each image layer into a preset human body identification model to obtain a human body part label corresponding to each image layer;
the fitting module is used for fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified;
the identification module is used for obtaining a human body position label corresponding to the target image sequence according to the human body position identification relation and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position identification relation and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
A computer device comprising a memory storing a computer program and a processor which when executing the computer program performs the steps of:
acquiring an image sequence to be identified, and extracting a preset number of image layers from the image sequence to be identified;
inputting each image layer into a preset human body identification model to obtain a human body part label corresponding to each image layer;
fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified;
obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:
acquiring an image sequence to be identified, and extracting a preset number of image layers from the image sequence to be identified;
inputting each image layer into a preset human body identification model to obtain a human body part label corresponding to each image layer;
fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified;
obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
According to the human body part identification method, the device, the computer equipment and the storage medium, the preset number of image layers are extracted from the image sequence to be identified, the extracted image layers are input into the preset human body identification model, the human body part labels corresponding to the image layers can be accurately obtained through the human body identification model, and further the position labels corresponding to the image layers and the human body part labels corresponding to the image layers can be accurately fitted by using the preset fitting method, so that the human body part identification relation can be accurately obtained, the human body part labels corresponding to the target image sequence can be accurately obtained according to the obtained human body part identification relation and the human body position labels corresponding to the target image sequence, or the human body position labels corresponding to the target image sequence can be accurately obtained according to the obtained human body part identification relation and the human body position labels corresponding to the target image sequence, and the accuracy of the human body part labels corresponding to the obtained target image sequence is improved; in addition, the preset number of image layers are extracted from the image sequence to be identified, so that resources consumed by calculating all the image layers can be saved, the calculation efficiency is improved, the human body part identification relationship can be obtained rapidly, the efficiency of obtaining the human body part identification relationship is improved, the application scene can be wider by extracting the preset number of image layers from the image sequence to be identified, and the application scene of the obtained human body part identification relationship is enlarged.
Drawings
FIG. 1 is a diagram of an application environment for a human body part recognition method in one embodiment;
FIG. 2 is a flow chart of a method for identifying a human body part according to an embodiment;
FIG. 2a is a schematic diagram illustrating an application of a method for identifying a human body part according to an embodiment;
FIG. 2b is a schematic diagram illustrating an application of a method for identifying a human body part according to an embodiment;
FIG. 2c is a schematic diagram illustrating an application of a method for identifying a human body part according to an embodiment;
FIG. 3 is a flowchart of a method for identifying a human body part according to another embodiment;
FIG. 4 is a flowchart of a method for identifying a human body part according to another embodiment;
FIG. 5 is a schematic diagram illustrating an application of a method for identifying a human body part according to an embodiment;
fig. 6 is a schematic structural diagram of a human body part recognition device according to an embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The human body part identification method provided by the embodiment of the application can be applied to the computer equipment shown in figure 1. The computer device comprises a processor, a memory, and a computer program stored in the memory, wherein the processor is connected through a system bus, and when executing the computer program, the processor can execute the steps of the method embodiments described below. Optionally, the computer device may further comprise a network interface, a display screen and an input means. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, which stores an operating system and a computer program, an internal memory. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external terminal through a network connection. Optionally, the computer device may be a server, a personal computer, a personal digital assistant, other terminal devices, such as a tablet computer, a mobile phone, etc., or a cloud or remote server, and the embodiment of the present application does not limit a specific form of the computer device.
There are two main ways of clinically identifying medical image content: firstly, the human body part mainly covered by the medical image is judged by combining the information such as research description (Study Description) and sequence description (Series Description) in the medical digital imaging and communication (Digital Imaging and Communications in Medicine, DICOM) information. The method can quickly and conveniently acquire the information of the human body part, but needs to be established under the condition that DICOM information is not wrong, and the condition that the DICOM information is filled by mistake occurs; in addition, only information of one part of the continuous scanning data can be acquired in the mode, for example, the continuous scanning data of the chest and the abdomen can only be read to one of the chest or the abdomen in DICOM information; secondly, when the DICOM information is filled or not fully filled, the actual part covered by the medical image data is checked manually, but the method is time-consuming. At present, there is also a method for identifying medical image content by combining feature extraction and a machine learning algorithm, mainly by extracting features of a medical image to be analyzed and identifying a human body part covered by the medical image to be analyzed through a classifier; or, the human body part is divided into a preset number of areas according to the specific anatomical structure of the human body part, numbering is carried out, and then the human body part covered by the medical image to be analyzed is identified by adopting a neural network method, but the problem of low identification accuracy exists. Accordingly, the present application provides a human body part recognition method, a computer device, and a storage medium capable of improving accuracy of human body part recognition.
In one embodiment, as shown in fig. 2, a human body part recognition method is provided, and the method is applied to the computer device in fig. 1 for illustration, and includes the following steps:
s201, acquiring an image sequence to be identified, and extracting a preset number of image layers from the image sequence to be identified.
The image sequence to be identified may be a DICOM sequence or a volume data sequence, and it should be noted that the DICOM sequence and the volume data sequence are identical in content, and the DICOM sequence may be converted into the volume data sequence.
Specifically, the computer device acquires an image sequence to be identified, and extracts a preset number of image layers from the image sequence to be identified. Alternatively, the computer device may obtain the image sequence to be identified from a PACS (Picture Archiving and Communication Systems, image archiving and communication system) server, or may obtain the image sequence to be identified from the image acquisition device in real time. Alternatively, the image sequence to be identified may be a computed tomography (Computed Tomography, CT) image sequence, a magnetic resonance (Magnetic Resonance Imaging, MRI) image sequence, or a low dose positron emission computed tomography (Positron Emission Computed Tomography/Magnetic Resonance Imaging, PET) image sequence. It can be understood that the image sequence to be identified includes a plurality of image layers, the computer device can extract a preset number of image layers from the plurality of image layers, and the time, the video memory, the memory and the like consumed by the computer device for calculating all the image layers in the image sequence to be identified can be saved by extracting the preset number of image layers.
S202, inputting each image layer into a preset human body recognition model to obtain a human body part label corresponding to each image layer.
Specifically, the computer equipment inputs the extracted preset number of image layers into a preset human body recognition model to obtain the human body part labels corresponding to the extracted image layers. The preset human body recognition model is a model which is trained in advance and used for recognizing the human body part labels corresponding to the extracted image layers. Alternatively, the network structure of the human body recognition model may be any network structure of a VGG network, a res net network, or a densnet network. Optionally, the computer device may input each image layer into a preset human body recognition model to obtain a human body part label corresponding to each image layer, or input each image layer into a preset human body recognition model at the same time to obtain a human body part label corresponding to each image layer. The body part label corresponding to each image layer may be a head, a chest or an abdomen, for example.
S203, fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified.
Specifically, the computer device utilizes a preset fitting method to fit the position labels corresponding to the image layers and the human body part labels corresponding to the image layers, so as to obtain a human body part identification relationship. The position label corresponding to each image layer is the position information of each image layer in the image sequence to be identified, and for example, the human body part label corresponding to each image layer may be any real value in the interval of 1-30, and the position label corresponding to each image layer may be any integer value in the interval of 1-100. Optionally, the fitting methodThe method comprises a linear fitting method and a nonlinear fitting method, namely, the computer equipment can fit the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset linear fitting method to obtain the human body part identification relationship, and can fit the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset nonlinear fitting method. Optionally, the preset linear fitting method may be any one algorithm of a gradient descent method, a least square method and a random sampling coincidence algorithm, or may be other fitting algorithms, and the linear fitting method is not limited in this embodiment. By way of example, the computer device may take the position label X for each image layer and the body part label Y for each image layer as two variables, fit the relationship F of X with respect to Y,wherein x is the position in the volume data or image sequence where the image layer to be identified is located,/->The human body part label of the image layer to be identified is predicted by fitting the relation F and the input x; conversely, the relation F of Y with respect to X can also be obtained -1 ，Wherein y is a human body part label corresponding to the image layer to be identified, and +.>Is obtained by fitting relation F -1 The human body position label of the image layer to be identified is predicted by the human body position label y corresponding to the image layer to be identified. Optionally, the preset fitting method may be any one algorithm of a gradient descent method, a least square method and a random sampling coincidence algorithm. The obtained human body part recognition relationship is the whole human body part recognition relationship.
S204, obtaining a human body position label corresponding to the target image sequence according to the human body position recognition relation and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position recognition relation and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
Specifically, the computer equipment obtains a human body part label corresponding to the target image sequence according to the obtained human body part recognition relationship, or obtains a human body position label corresponding to the target image sequence according to the obtained human body part recognition relationship, wherein the target image sequence is any sequence in the image sequence to be recognized. Illustratively, the computer device can identify the relationship based on the human body parts obtained aboveObtaining a human body part label corresponding to the target image sequence, and identifying the relationship ∈10 according to the obtained human body part>And obtaining a human body position label corresponding to the target image sequence.
After obtaining the human body part recognition relationship, in some scenes, some applications can be performed according to the obtained human body part recognition relationship, 1) as shown in fig. 2a, the computer equipment can refer to the human anatomy structure, divide the human body part into N areas and number the N areas (combined with the piecewise linear distribution of human anatomy mark points), then recognize the image through a convolutional neural network or an image recognition algorithm, directly detect the single-layer image, extract a certain number of single-layer images from the volume data, and infer (post-process and judge) the human body part label and the position label by combining the obtained human body part recognition relationship, thereby obtaining the occupation ratio of the human body part contained in the volume data; 2) As shown in fig. 2b, the computer device may obtain a position tag corresponding to the original image data by using the obtained human body part recognition relationship through the human body part automatic recognition module, match the position tag of the obtained image data with a related part algorithm module in the algorithm library, and call an algorithm to perform subsequent image analysis and computer-aided diagnosis to obtain a final processing result; 3) As shown in fig. 2c, the automatic human body part recognition module performs part recognition on the image data in the image database by using the obtained human body part recognition relationship, and performs subsequent data screening, data processing and other tasks after obtaining the position tag and the human body part tag corresponding to the image data.
In the human body part recognition method, the computer equipment extracts the preset number of image layers from the image sequence to be recognized, inputs the extracted image layers into the preset human body recognition model, can accurately obtain the human body part label corresponding to each image layer through the human body recognition model, further can accurately fit the position label corresponding to each image layer and the human body part label corresponding to each image layer by using the preset fitting method, and accurately obtain the human body part recognition relation, so that the human body part label corresponding to the target image sequence can be accurately obtained according to the obtained human body part recognition relation and the human body position label corresponding to the target image sequence, or the human body position label corresponding to the target image sequence can be accurately obtained according to the obtained human body part recognition relation and the human body part label corresponding to the target image sequence, and the accuracy of the human body part label corresponding to the obtained target image sequence or the human body position label corresponding to the target image sequence is improved; in addition, the preset number of image layers are extracted from the image sequence to be identified, so that resources consumed by calculating all the image layers can be saved, the calculation efficiency is improved, the human body part identification relationship can be obtained rapidly, the efficiency of obtaining the human body part identification relationship is improved, the application scene can be wider by extracting the preset number of image layers from the image sequence to be identified, and the application scene of the obtained human body part identification relationship is enlarged.
And obtaining the human body position label corresponding to the target image sequence in the scene according to the human body position identification relation and the human body position label corresponding to the target image sequence. In one embodiment, as shown in fig. 3, S204 includes:
s301, acquiring a human body position label corresponding to the target image sequence.
Specifically, the computer device acquires a human body position tag corresponding to the target image sequence. Optionally, the computer device may obtain a human body position tag corresponding to the target image sequence according to the position information of the target image sequence in the image sequence to be identified, that is, determine the position information of the target image sequence in the image sequence to be identified as the human body position tag corresponding to the target image sequence. It can be understood that the target image sequence is any sequence in the image sequence to be identified, and when the target image sequence is acquired from the image sequence to be identified, the position information of the target image sequence in the image sequence to be identified is obtained, so that the human body position label corresponding to the target image sequence can be correspondingly acquired.
S302, predicting the human body part label corresponding to the target image sequence according to the human body position label corresponding to the target image sequence and the human body part recognition relation to obtain the human body part label of the target image sequence.
Specifically, the computer device predicts the human body part label corresponding to the target image sequence according to the human body position label corresponding to the target image sequence and the obtained human body part recognition relationship, and obtains the human body part label of the target image sequence. The computer device can identify the relationship between the human body position label corresponding to the target image sequence and the human body partAnd predicting the human body part label corresponding to the target image sequence to obtain the human body part label of the target image sequence.
In this embodiment, the computer device may accurately predict the human body position label corresponding to the target image sequence according to the human body position label corresponding to the obtained target image sequence and the human body position identification relationship by obtaining the human body position label corresponding to the target image sequence, thereby improving the accuracy of the human body position label of the obtained target image sequence.
And obtaining the human body position label corresponding to the target image sequence in the scene according to the human body position identification relation and the human body position label corresponding to the target image sequence. In one embodiment, as shown in fig. 4, S204 includes:
s401, acquiring a human body part label corresponding to a target image sequence; the label of the human body part corresponding to the target image sequence is the label of the human body part to which the target image sequence belongs.
Specifically, the computer device first acquires a human body part label corresponding to the target image sequence. The human body part label corresponding to the target image sequence is a label of a human body part to which the target image sequence belongs. Optionally, the computer device may obtain the body part label corresponding to the target image sequence according to the body part label sequence to which the target image sequence belongs. I.e. the computer device first determines whether the target image sequence corresponds to a sequence of images of the head of a human body, a sequence of images of the chest of a human body, or a sequence of images of the abdomen of a human body.
S402, predicting the position label corresponding to the target image sequence according to the human body part label corresponding to the target image sequence and the human body part recognition relation to obtain the position label corresponding to the target image sequence.
Specifically, the computer device predicts the position label corresponding to the target image sequence according to the obtained human body part label corresponding to the target image sequence and the human body part identification relationship, and obtains the position label corresponding to the target image sequence. The computer device may, for example, identify the relationship between the human body part label corresponding to the target image sequence and the human body partAnd predicting the position label corresponding to the target image sequence to obtain the position label corresponding to the target image sequence.
In this embodiment, the computer device may accurately predict the position tag corresponding to the target image sequence by acquiring the human body part tag corresponding to the target image sequence according to the human body part tag corresponding to the acquired target image sequence and the human body part recognition relationship, so as to improve the accuracy of the position tag corresponding to the acquired target image sequence.
In the above scenario of extracting the preset number of image layers from the image sequence to be identified, the computer device may randomly extract the preset number of image layers, or may equally extract the preset number of image layers. In one embodiment, the step S201 includes: randomly extracting the preset number of image layers from the image sequence to be identified, or equally extracting the preset number of image layers from the image sequence to be identified according to a preset extraction interval.
Specifically, the computer device randomly extracts the preset number of images from the image sequence to be identified according to the preset number, or equally extracts the preset number of image layers from the image sequence to be identified according to a preset extraction interval. For example, if the preset number of image layers is 5 image layers, the computer device may randomly extract 5 images from the image sequence to be identified to obtain the preset number of image layers, or may extract image layers from each 4 image layers in the image sequence to be identified, so as to obtain the 5 image layers.
In this embodiment, the computer device randomly extracts a preset number of image layers from the image sequence to be identified, or equally extracts a preset number of image layers from the image sequence to be identified according to a preset extraction interval, so that resources consumed in calculating all the image layers can be saved, the calculation efficiency is improved, and the application scene can be wider by extracting the preset number of image layers from the image sequence to be identified, so that the application scene of the obtained human body part identification relationship is enlarged.
In some scenarios, for example, as shown in fig. 5, in a scenario of the key organ matching workflow, after obtaining the body part tag or the body position corresponding to the target image sequence, the body organ to which the target image sequence belongs needs to be obtained, and in one embodiment, the method further includes: and obtaining the human organ to which the target image sequence belongs according to the mapping relation between the preset human organ and the human body part label or the mapping relation between the preset human organ and the human body position label.
Specifically, the computer device obtains the human organ to which the target image sequence belongs according to the mapping relation between the preset human organ and the human body part label, or obtains the human organ to which the target image sequence belongs according to the mapping relation between the preset human organ and the human body position label. If the label of the target image sequence obtained by the computer equipment is a label of the head, the computer equipment obtains the human organ to which the target image sequence belongs as the head according to the preset mapping relation between the human organ and the label of the human part; if the human body position label of the target image sequence obtained by the computer equipment is 10, the computer equipment can obtain the human body organ to which the target image sequence belongs according to the preset mapping relation between the human body organ and the human body position label.
In this embodiment, the computer device may quickly obtain the human organ to which the target image sequence belongs according to the mapping relationship between the preset human organ and the human body part label, or the mapping relationship between the preset human organ and the human body position label, so as to improve the efficiency of determining the human organ to which the target image sequence belongs.
It should be understood that, although the steps in the flowcharts of fig. 2-4 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2-4 may include multiple steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the steps or stages in other steps or other steps.
In one embodiment, as shown in fig. 6, there is provided a human body part recognition apparatus comprising: the device comprises a first acquisition module, a second acquisition module, a fitting module and an identification module, wherein:
the first acquisition module is used for acquiring the image sequence to be identified and extracting a preset number of image layers from the image sequence to be identified.
The second acquisition module is used for inputting each image layer into a preset human body recognition model to obtain a human body part label corresponding to each image layer.
The fitting module is used for fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified.
The identification module is used for obtaining a human body position label corresponding to the target image sequence according to the human body position identification relation and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position identification relation and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
Optionally, the fitting method includes a linear fitting method or a nonlinear fitting method.
The human body part recognition device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the above identification module includes a first acquisition unit and a first prediction unit, where:
the first acquisition unit is used for acquiring the human body position label corresponding to the target image sequence.
The first prediction unit is used for predicting the human body part label corresponding to the target image sequence according to the human body position label corresponding to the target image sequence and the human body part recognition relation to obtain the human body part label of the target image sequence.
The human body part recognition device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the first obtaining unit is specifically configured to obtain a human body position tag corresponding to the target image sequence according to position information of the target image sequence in the image sequence to be identified.
The human body part recognition device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the above identification module includes a second acquisition unit and a second prediction unit, where:
the second acquisition unit is used for acquiring a human body part label corresponding to the target image sequence; the label of the human body part corresponding to the target image sequence is the label of the human body part to which the target image sequence belongs.
The second prediction unit is used for predicting the position label corresponding to the target image sequence according to the human body part label corresponding to the target image sequence and the human body part identification relationship to obtain the position label corresponding to the target image sequence.
The human body part recognition device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the second obtaining unit is specifically configured to obtain a body part label corresponding to the target image sequence according to a body part label sequence to which the target image sequence belongs.
The human body part recognition device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the first obtaining module includes an extracting unit, where:
the extraction unit is used for randomly extracting the preset number of image layers from the image sequence to be identified, or equally extracting the preset number of image layers from the image sequence to be identified according to preset extraction intervals.
The human body part recognition device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
On the basis of the above embodiment, optionally, the above apparatus further includes: a third acquisition module, wherein:
the third acquisition module is used for obtaining the human organ to which the target image sequence belongs according to the mapping relation between the preset human organ and the human body part label or the mapping relation between the preset human organ and the human body position label.
The human body part recognition device provided in this embodiment may perform the above method embodiment, and its implementation principle and technical effects are similar, and will not be described herein.
For specific limitations of the human body part recognition apparatus, reference may be made to the above limitations of the human body part recognition method, and no further description is given here. The above-described individual modules in the human body part recognition apparatus may be implemented in whole or in part by software, hardware, and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:
acquiring an image sequence to be identified, and extracting a preset number of image layers from the image sequence to be identified;
inputting each image layer into a preset human body identification model to obtain a human body part label corresponding to each image layer;
fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified;
obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
The computer device provided in the foregoing embodiments has similar implementation principles and technical effects to those of the foregoing method embodiments, and will not be described herein in detail.
In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring an image sequence to be identified, and extracting a preset number of image layers from the image sequence to be identified;
inputting each image layer into a preset human body identification model to obtain a human body part label corresponding to each image layer;
fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified;
obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
The computer readable storage medium provided in the above embodiment has similar principle and technical effects to those of the above method embodiment, and will not be described herein.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (10)
1. A method of identifying a body part, the method comprising:
acquiring an image sequence to be identified, and extracting a preset number of image layers from the image sequence to be identified;
inputting each image layer into a preset human body identification model to obtain a human body part label corresponding to each image layer;
fitting the position labels corresponding to the image layers and the human body part labels corresponding to the image layers by using a preset fitting method to obtain a human body part identification relationship; the position labels corresponding to the image layers are the position information of the image layers in the image sequence to be identified;
obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence, or obtaining a human body position label corresponding to the target image sequence according to the human body position identification relationship and the human body position label corresponding to the target image sequence; the target image sequence is any sequence in the image sequence to be identified.
2. The method according to claim 1, wherein the obtaining the human body part label corresponding to the target image sequence according to the human body part identification relationship and the human body position label corresponding to the target image sequence includes:
acquiring a human body position label corresponding to the target image sequence;
and predicting the human body part label corresponding to the target image sequence according to the human body position label corresponding to the target image sequence and the human body part identification relationship to obtain the human body part label of the target image sequence.
3. The method according to claim 2, wherein the acquiring the human body position tag corresponding to the target image sequence includes:
and obtaining a human body position label corresponding to the target image sequence according to the position information of the target image sequence in the image sequence to be identified.
4. The method according to claim 1, wherein the obtaining the human body position tag corresponding to the target image sequence according to the human body position recognition relationship and the human body position tag corresponding to the target image sequence includes:
acquiring a human body part label corresponding to the target image sequence; the label of the human body part corresponding to the target image sequence is the label of the human body part to which the target image sequence belongs;
and predicting the position label corresponding to the target image sequence according to the human body part label corresponding to the target image sequence and the human body part identification relationship to obtain the position label corresponding to the target image sequence.
5. The method of claim 4, wherein the acquiring the body part tag corresponding to the target image sequence comprises:
and obtaining a human body part label corresponding to the target image sequence according to the human body part label sequence to which the target image sequence belongs.
6. The method of claim 1, wherein the fitting method comprises a linear fitting method or a nonlinear fitting method.
7. The method according to any of claims 1-6, wherein said extracting a preset number of image layers from said sequence of images to be identified comprises:
randomly extracting the preset number of image layers from the image sequence to be identified, or equally extracting the preset number of image layers from the image sequence to be identified according to a preset extraction interval.
8. The method according to claim 1, wherein the method further comprises:
and obtaining the human organ to which the target image sequence belongs according to the mapping relation between the preset human organ and the human body part label or the mapping relation between the preset human organ and the human body position label.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 8 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010787578.1A CN112102235B (en) | 2020-08-07 | 2020-08-07 | Human body part recognition method, computer device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010787578.1A CN112102235B (en) | 2020-08-07 | 2020-08-07 | Human body part recognition method, computer device, and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112102235A CN112102235A (en) | 2020-12-18 |
| CN112102235B true CN112102235B (en) | 2023-10-27 |
Family
ID=73749979
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010787578.1A Active CN112102235B (en) | 2020-08-07 | 2020-08-07 | Human body part recognition method, computer device, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112102235B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112614568B (en) * | 2020-12-28 | 2024-05-28 | 东软集团股份有限公司 | Method and device for processing inspection image, storage medium and electronic equipment |
| CN112766314B (en) * | 2020-12-31 | 2024-05-28 | 上海联影智能医疗科技有限公司 | Anatomical structure recognition method, electronic device, and storage medium |
| CN113177923B (en) * | 2021-05-07 | 2024-06-21 | 上海联影智能医疗科技有限公司 | Medical image content identification method, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016147545A1 (en) * | 2015-03-19 | 2016-09-22 | パナソニックＩｐマネジメント株式会社 | Image display device, image display system, image display method, and program |
| CN107423721A (en) * | 2017-08-08 | 2017-12-01 | 珠海习悦信息技术有限公司 | Interactive action detection method, device, storage medium and processor |
| CN111027469A (en) * | 2019-12-09 | 2020-04-17 | 上海联影智能医疗科技有限公司 | Human body part recognition method, computer device and readable storage medium |
| WO2020119527A1 (en) * | 2018-12-11 | 2020-06-18 | 中国科学院深圳先进技术研究院 | Human action recognition method and apparatus, and terminal device and storage medium |
-
2020
- 2020-08-07 CN CN202010787578.1A patent/CN112102235B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016147545A1 (en) * | 2015-03-19 | 2016-09-22 | パナソニックＩｐマネジメント株式会社 | Image display device, image display system, image display method, and program |
| CN107423721A (en) * | 2017-08-08 | 2017-12-01 | 珠海习悦信息技术有限公司 | Interactive action detection method, device, storage medium and processor |
| WO2020119527A1 (en) * | 2018-12-11 | 2020-06-18 | 中国科学院深圳先进技术研究院 | Human action recognition method and apparatus, and terminal device and storage medium |
| CN111027469A (en) * | 2019-12-09 | 2020-04-17 | 上海联影智能医疗科技有限公司 | Human body part recognition method, computer device and readable storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 产品表面缺陷检测的变步长采样机制研究;郭静;韩跃平;李会鸽;;科技通报(02);全文 * |
| 行人再识别技术综述;李幼蛟;卓力;张菁;李嘉锋;张辉;;自动化学报(09);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112102235A (en) | 2020-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109872306B (en) | Medical image segmentation method, device and storage medium | |
| US10810735B2 (en) | Method and apparatus for analyzing medical image | |
| CN112102235B (en) | Human body part recognition method, computer device, and storage medium | |
| CN111160367B (en) | Image classification method, apparatus, computer device, and readable storage medium | |
| CN111950329A (en) | Target detection and model training method and device, computer equipment and storage medium | |
| CN110781859B (en) | Image annotation method and device, computer equipment and storage medium | |
| CN111369542A (en) | Blood vessel marking method, image processing system and storage medium | |
| CN111583199B (en) | Sample image labeling method, device, computer equipment and storage medium | |
| US10740901B2 (en) | Encoder regularization of a segmentation model | |
| CN110210519B (en) | Classification method, computer device, and storage medium | |
| CN111210465B (en) | Image registration method, image registration device, computer equipment and readable storage medium | |
| WO2022057309A1 (en) | Lung feature recognition method and apparatus, computer device, and storage medium | |
| CN111583184A (en) | Image analysis method, network, computer device, and storage medium | |
| CN111080583B (en) | Medical image detection method, computer device, and readable storage medium | |
| CN111325714A (en) | Region-of-interest processing method, computer device and readable storage medium | |
| KR101885562B1 (en) | Method for mapping region of interest in first medical image onto second medical image and apparatus using the same | |
| CN115861248A (en) | Medical image segmentation method, medical model training method, medical image segmentation device and storage medium | |
| CN113160199B (en) | Image recognition method and device, computer equipment and storage medium | |
| CN113724185B (en) | Model processing method, device and storage medium for image classification | |
| CN111462203B (en) | DR focus evolution analysis device and method | |
| CN111027469B (en) | Human body part recognition method, computer device, and readable storage medium | |
| CN111209946B (en) | Three-dimensional image processing method, image processing model training method and medium | |
| CN110738175B (en) | Face image processing method and device, computer equipment and storage medium | |
| CN113177923B (en) | Medical image content identification method, electronic equipment and storage medium | |
| CN114724016A (en) | Image classification method, computer device, and storage medium |
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 |