CN113160144B - Target object detection method, target object detection device, electronic equipment and storage medium - Google Patents
Target object detection method, target object detection device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113160144B CN113160144B CN202110325008.5A CN202110325008A CN113160144B CN 113160144 B CN113160144 B CN 113160144B CN 202110325008 A CN202110325008 A CN 202110325008A CN 113160144 B CN113160144 B CN 113160144B
- Authority
- CN
- China
- Prior art keywords
- target object
- detection
- image
- confidence
- standard
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 274
- 238000004458 analytical method Methods 0.000 claims abstract description 38
- 230000007246 mechanism Effects 0.000 claims abstract description 17
- 238000012216 screening Methods 0.000 claims abstract description 11
- 238000007689 inspection Methods 0.000 claims description 29
- 238000000034 method Methods 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 206010028980 Neoplasm Diseases 0.000 description 23
- 230000006870 function Effects 0.000 description 16
- 238000011176 pooling Methods 0.000 description 10
- 238000007726 management method Methods 0.000 description 5
- 238000010606 normalization Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 102100025444 Gamma-butyrobetaine dioxygenase Human genes 0.000 description 2
- 101000934612 Homo sapiens Gamma-butyrobetaine dioxygenase Proteins 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- 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
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Radiology & Medical Imaging (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Probability & Statistics with Applications (AREA)
- Public Health (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to artificial intelligence technology, and discloses a target object detection method, which comprises the following steps: performing coarse positioning on an original image set to obtain target positioning confidence of each image in the original image set, performing confidence analysis according to the target positioning confidence, selecting a plurality of images from the original image set as standard images, performing target initial detection on the standard images to obtain initial detection scores of targets, selecting candidate images from the standard images according to the initial detection scores and the target positioning confidence, performing target final detection on the candidate images, and screening out final targets from target final detection results by using a preset voting mechanism. Furthermore, the present invention relates to blockchain technology, and the final object can be stored in a node of the blockchain. The invention also provides a target object detection device, electronic equipment and a computer readable storage medium. The invention can solve the problem of lower detection accuracy of the target object.
Description
Technical Field
The present invention relates to the field of artificial intelligence technologies, and in particular, to a target object detection method, a target object detection device, an electronic device, and a computer readable storage medium.
Background
Object detection has wide application in various fields. For example, in tumor detection, localization of tumors in medical images is an important part of cancer examination and clinical treatment procedures. The existing method is to use one detection network to locate all tumors in 2D, however, in clinical treatment, only the main tumor needs to be found, the main layer where the main tumor is located (key ROI), and the clinical operation of completing tumor location by using a single detection network leads to lower tumor identification accuracy.
Disclosure of Invention
The invention provides a target object detection method, a target object detection device and a computer readable storage medium, and mainly aims to solve the problem of low target object detection accuracy.
In order to achieve the above object, the present invention provides a target object detection method, including:
performing coarse positioning of a target object on an original image set by using a pre-constructed image classification network, and obtaining the positioning confidence coefficient of the target object of each image in the original image set;
performing confidence analysis on the original image set according to the target object positioning confidence, and selecting a plurality of images from the original image set to serve as standard images based on the confidence analysis;
Performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, and calculating the primary detection score of the target object based on the primary detection result of the target object;
selecting candidate images from the standard images according to the initial detection score and the target object positioning confidence;
and carrying out target object final inspection on the candidate images by utilizing the image detection network to obtain a target object final inspection result, and screening out a final target object from the target object final inspection result by utilizing a preset voting mechanism.
Optionally, the performing coarse positioning of the target object on the original image set by using the pre-constructed image classification network, and obtaining the target object positioning confidence coefficient of each image in the original image set includes:
layering the obtained target object image to obtain an original image set consisting of a plurality of image layers;
and performing classification prediction on each image in the original image set layer by utilizing the pre-constructed image classification network to obtain the target object positioning confidence of each image.
Optionally, the performing confidence analysis on the original image set according to the object positioning confidence, selecting a plurality of images from the original image set as standard images based on the confidence analysis, including:
Constructing a coordinate axis by utilizing the layer direction of the image layer and the target object positioning confidence, and drawing a positioning confidence curve in the coordinate axis;
calculating a half-width value corresponding to each peak value in the position credibility curve;
and taking the original image layer with the half-width value as a starting layer and an ending layer, and acquiring the image layer between the starting layer and the ending layer as the standard image.
Optionally, the performing the primary detection of the target object on the standard image by using the pre-constructed image detection network to obtain a primary detection result of the target object, and calculating the primary detection score of the target object based on the primary detection result of the target object, including:
performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, wherein the primary detection result of the target object comprises a candidate frame and confidence corresponding to the candidate frame;
and carrying out normalization processing on the candidate frames, and obtaining the primary detection score of the standard image based on the confidence coefficient corresponding to the candidate frames.
Optionally, the normalizing the candidate frame, and obtaining the initial detection score of the standard image based on the confidence coefficient corresponding to the candidate frame includes:
The preliminary score was calculated using the following formula:
wherein ,
for the primary check score of the mth standard image, < >>
Confidence of kth candidate frame for mth standard image, a m,k A value in the interval 0 to 1, +.>
The number of candidate frames in the mth standard image.
Optionally, the selecting a candidate image from the standard image according to the initial detection score and the target object positioning confidence comprises:
adding and sequencing the initial detection scores of the standard images and the target object positioning confidence corresponding to the standard images;
and selecting a preset number of standard images from high to low as candidate images based on the added scores.
Optionally, the screening the final target object from the target object final detection result by using a preset voting mechanism includes:
the target object final detection result comprises detection frames and confidence degrees corresponding to the detection frames, pixel points with the highest confidence degrees in each detection frame are selected, and confidence average calculation is carried out on all detection frames containing the pixel points to obtain final detection scores;
and selecting the target object in the detection frame with the highest final detection score as the final target object.
In order to solve the above problems, the present invention also provides an object detection apparatus, including:
the target object coarse positioning module is used for performing target object coarse positioning on an original image set by utilizing a pre-constructed image classification network, and obtaining the target object positioning confidence coefficient of each image in the original image set;
the confidence analysis module is used for carrying out confidence analysis on the original image set according to the object positioning confidence, and selecting a plurality of images from the original image set to serve as standard images based on the confidence analysis;
the target object primary detection module is used for carrying out target object primary detection on the standard image by utilizing a pre-constructed image detection network to obtain a target object primary detection result, and calculating the primary detection score of the target object based on the target object primary detection result;
the image selecting module is used for selecting candidate images from the standard images according to the initial detection score and the target object positioning confidence;
and the target object final inspection module is used for performing target object final inspection on the candidate images by utilizing the image detection network to obtain target object final inspection results, and screening out final target objects from the target object final inspection results by utilizing a preset voting mechanism.
In order to solve the above-mentioned problems, the present invention also provides an electronic apparatus including:
a memory storing at least one instruction; and
And the processor executes the instructions stored in the memory to realize the target object detection method.
In order to solve the above-mentioned problems, the present invention also provides a computer-readable storage medium having stored therein at least one instruction that is executed by a processor in an electronic device to implement the above-mentioned object detection method.
According to the method, the pre-constructed image classification network is utilized to perform coarse positioning of the target object on the original image set, the target object positioning confidence coefficient of each image in the original image set is obtained, confidence analysis is performed on the original image set based on the target object positioning confidence coefficient, a standard image is obtained through a confidence coefficient curve obtained through the confidence analysis, and the accuracy of target object detection can be improved. And the pre-constructed image detection network is utilized to perform initial detection and final detection of the target object on the standard image, so that the accuracy of target object detection is further improved. Meanwhile, the detection result is optimized based on a voting mechanism, a final target object is screened out from the final detection result of the target object, a large number of detection results are not output, and the efficiency of target object detection is improved. Therefore, the target object detection method, the target object detection device, the electronic equipment and the computer readable storage medium can solve the problem of low target object detection accuracy.
Drawings
FIG. 1 is a flow chart of a method for detecting a target object according to an embodiment of the invention;
FIG. 2 is a detailed flow chart of one of the steps shown in FIG. 1;
FIG. 3 is a detailed flow chart of another step of FIG. 1;
FIG. 4 is a detailed flow chart of another step of FIG. 1;
FIG. 5 is a detailed flow chart of another step of FIG. 1;
FIG. 6 is a functional block diagram of a target detection apparatus according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of an electronic device for implementing the target object detection method according to an embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
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 invention.
The embodiment of the application provides a target object detection method. The execution subject of the target object detection method includes, but is not limited to, at least one of a server, a terminal, and the like, which can be configured to execute the method provided in the embodiments of the present application. In other words, the object detection method may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like.
Referring to fig. 1, a flow chart of a target object detection method according to an embodiment of the invention is shown.
In this embodiment, the target object detection method includes:
s1, performing coarse positioning on an original image set by utilizing a pre-constructed image classification network, and obtaining the target positioning confidence of each image in the original image set.
Specifically, referring to fig. 2, the coarse positioning of the object in the original image set by using the pre-constructed image classification network, and obtaining the positioning confidence of the object in each image in the original image set, includes:
s10, layering the obtained target object image to obtain an original image set consisting of a plurality of image layers;
s11, performing classification prediction on each image in the original image set layer by utilizing the pre-constructed image classification network to obtain the target object positioning confidence coefficient of each image.
The target image may be a three-dimensional medical image of a tumor, the pre-constructed image classification network may be VGGNet, where VGGNet is one of deep convolutional neural networks, and compared with a general neural network, VGGNet uses a convolutional kernel of 3*3 and a pooling kernel of 2×2, and performance is improved by continuously deepening a network structure, and meanwhile, an increase in the number of network layers does not bring about explosion in the number of parameters, because parameter amounts are mainly concentrated in the last three fully-connected layers. The VGGNet can improve the classification effect of the original image set.
Taking a three-dimensional medical image of a tumor as an example, layering an original image of the tumor (one or more original images are arranged on each layer) to obtain an original image set composed of a plurality of image layers, inputting the original image set into the image classification network, performing two classifications on the original images in the original image set by using the image classification network (namely, predicting the confidence of the tumor contained in the original image of each layer), outputting the classification confidence of each layer, and recording as
Representing the classification confidence of the m-th layer.
S2, carrying out confidence analysis on the original image set according to the target object positioning confidence, and selecting a plurality of images from the original image set to serve as standard images based on the confidence analysis.
Specifically, referring to fig. 3, the performing a confidence analysis on the original image set according to the object positioning confidence, and selecting a plurality of images from the original image set as standard images based on the confidence analysis includes:
s20, constructing a coordinate axis by using the layer direction of the image layer and the target object positioning confidence, and drawing a positioning confidence curve in the coordinate axis;
S21, calculating a half-width value corresponding to each peak value in the positioning confidence coefficient curve;
s22, taking the original image layer with the half-width value as a starting layer and an ending layer, and acquiring the image layer between the starting layer and the ending layer as the standard image.
In the embodiment of the present invention, the layer direction of the image layer may be taken as the x-axis, and the confidence level may be located by the target object
For the y-axis, a plot of confidence as a function of z-direction is drawn. Detection of tumorsFor example, since the original images in the original image set do not necessarily have tumors, a plurality of standard images containing tumors can be output by drawing the target object position credibility curve, so that the detection precision is improved.
In the embodiment of the invention, the confidence degree analysis is carried out on the original image set through the target object positioning confidence degree, and the standard image is obtained through the confidence degree curve obtained through the confidence degree analysis, so that the accuracy of target object detection can be improved.
S3, performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, and calculating the primary detection score of the target object based on the primary detection result of the target object.
In the embodiment of the present invention, the pre-constructed image detection network may be a fast RCNN detection model, where the fast RCNN detection model includes a conversion layer (Conv filters), a RPN (Region Proposal Networks) network, a Pooling layer (Roi Pooling), and a Classification layer (Classification). The transformation layer (Conv layers) transforms the standard image into a feature map by convolution, which is a linear operation, and the convolution operation of the image can eliminate noise and enhance features. The RPN network comprises anchor frame generation, detection function judgment and frame regression, a candidate region is obtained by generating a real boundary frame and a series of anchor frames in the feature map, whether a target exists in the candidate region is judged by using a detection function, and frame regression is carried out on the candidate region with the target, so that an accurate candidate region (namely a region in the candidate frame) is obtained. The anchor frame is a prediction boundary frame generated by taking pixel points as centers and collecting a large number of areas in the characteristic map. The detection function may use a softmax function. The frame regression refers to a process of approaching the generated prediction boundary frame with the marked real boundary frame as a target in the target detection process. The exact candidate regions may vary in size, and the exact candidate region is converted to a fixed-size image using the Pooling layer (Roi Pooling). The Classification layer (Classification) is used to determine which category (e.g., whether a tumor) the target in the precise candidate region specifically belongs to and to output a primary score for each primary layer.
Specifically, referring to fig. 4, the performing, by using a pre-constructed image detection network, initial detection of the target object on the standard image to obtain an initial detection result of the target object, and calculating an initial detection score of the target object based on the initial detection result of the target object includes:
s30, performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, wherein the primary detection result of the target object comprises a candidate frame and confidence corresponding to the candidate frame;
s31, carrying out normalization processing on the candidate frames, and obtaining the initial detection score of the standard image based on the confidence coefficient corresponding to the candidate frames.
In the embodiment of the present invention, the normalizing the candidate frame and obtaining the initial detection score of the standard image based on the confidence coefficient corresponding to the candidate frame includes:
the preliminary score was calculated using the following formula:
wherein ,
for the primary check score of the mth standard image, < >>
Confidence of kth candidate frame for mth standard image, a m,k A value in the interval 0 to 1, +.>
The number of candidate frames in the mth standard image.
In the embodiment of the present invention, the normalization processing may be that the mth standard image, the kth candidate frame area is divided by all the candidate frame areas in the standard image. And the standard image is finely positioned by utilizing the pre-constructed image detection network, so that the accuracy of target object detection is further improved.
And S4, selecting candidate images from the standard images according to the initial detection score and the target object positioning confidence.
Specifically, referring to fig. 5, the S4 includes:
s40, adding and sequencing the initial detection scores of the standard images and the target object positioning confidence corresponding to the standard images;
s41, selecting a preset number of standard images from high to low based on the added scores as candidate images.
Wherein the detection score of each standard image is calculated by
And the corresponding target positioning confidence level->
Adding, wherein a preset number of detection layers is selected from the high score to the low score, and the preset number can be 40% of the number of the main layers.
S5, performing target object final inspection on the candidate images by using the image detection network to obtain target object final inspection results, and screening out final target objects from the target object final inspection results by using a preset voting mechanism.
In the embodiment of the invention, the pre-constructed image detection network may also be a fast RCNN detection model, and the final detection result of the target object includes a detection frame and a confidence coefficient corresponding to the detection frame.
Specifically, the screening the final target object from the final target object detection result by using a preset voting mechanism includes:
The target object final detection result comprises detection frames and confidence degrees corresponding to the detection frames, pixel points with the highest confidence degrees in each detection frame are selected, and confidence average calculation is carried out on all detection frames containing the pixel points to obtain final detection scores;
and selecting the target object in the detection frame with the highest final detection score as the final target object.
In the embodiment of the invention, two pixel points R are used 1 and R2 For example, assume that there are two detection boxes (BBOX) for each pixel, where R 1 The corresponding detection frames are respectively
The corresponding confidence is
The corresponding confidence is->
R 2 The corresponding detection frames are respectively
The corresponding confidence is->
The corresponding confidence is->
The final detection scores are +.>
And selecting the target object in the detection frame with the highest final detection score as the final detection result.
In the embodiment of the invention, the final detection result of the target object is optimized based on the voting mechanism, so that the final target object is obtained, and the target object detection efficiency is improved.
According to the method, the pre-constructed image classification network is utilized to perform coarse positioning of the target object on the original image set, the target object positioning confidence coefficient of each image in the original image set is obtained, confidence analysis is performed on the original image set based on the target object positioning confidence coefficient, a standard image is obtained through a confidence coefficient curve obtained through the confidence analysis, and the accuracy of target object detection can be improved. And the pre-constructed image detection network is utilized to perform initial detection and final detection of the target object on the standard image, so that the accuracy of target object detection is further improved. Meanwhile, the detection result is optimized based on a voting mechanism, a final target object is screened out from the final detection result of the target object, a large number of detection results are not output, and the efficiency of target object detection is improved. Therefore, the embodiment of the invention can solve the problem of lower detection accuracy of the target object.
Fig. 6 is a functional block diagram of an object detection device according to an embodiment of the present invention.
The object detection device 100 of the present invention may be mounted in an electronic apparatus. Depending on the implementation, the target detection apparatus 100 may include a target coarse positioning module 101, a confidence analysis module 102, a target primary detection module 103, an image selection module 104, and a target final detection module 105. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
the target object coarse positioning module 101 is configured to perform target object coarse positioning on an original image set by using a pre-constructed image classification network, and obtain a target object positioning confidence coefficient of each image in the original image set.
Specifically, the target coarse positioning module 101 obtains the target positioning confidence of each image in the original image set by:
layering the obtained target object image to obtain an original image set consisting of a plurality of image layers;
And performing classification prediction on each image in the original image set layer by utilizing the pre-constructed image classification network to obtain the target object positioning confidence of each image.
The target image may be a three-dimensional medical image of a tumor, the pre-constructed image classification network may be VGGNet, where VGGNet is one of deep convolutional neural networks, and compared with a general neural network, VGGNet uses a convolutional kernel of 3*3 and a pooling kernel of 2×2, and performance is improved by continuously deepening a network structure, and meanwhile, an increase in the number of network layers does not bring about explosion in the number of parameters, because parameter amounts are mainly concentrated in the last three fully-connected layers. The VGGNet can improve the classification effect of the original image set.
Taking a three-dimensional medical image of a tumor as an example, layering an original image of the tumor (one or more original images are arranged on each layer) to obtain an original image set composed of a plurality of image layers, inputting the original image set into the image classification network, performing two classifications on the original images in the original image set by using the image classification network (namely, predicting the confidence of the tumor contained in the original image of each layer), outputting the classification confidence of each layer, and recording as 
Representing the classification confidence of the m-th layer.
The confidence analysis module 102 is configured to perform confidence analysis on the original image set according to the object positioning confidence, and select a plurality of images from the original image set as standard images based on the confidence analysis.
Specifically, the confidence analysis module 102 selects several images from the original image set as standard images by:
constructing a coordinate axis by utilizing the layer direction of the image layer and the target object positioning confidence, and drawing a positioning confidence curve in the coordinate axis;
calculating a half-width value corresponding to each peak value in the position credibility curve;
and taking the original image layer with the half-width value as a starting layer and an ending layer, and acquiring the image layer between the starting layer and the ending layer as the standard image.
In the embodiment of the invention, the image can be usedThe layer direction of the layer is the x axis, and the object positioning confidence is used
For the y-axis, a plot of confidence as a function of z-direction is drawn. Taking tumor detection as an example, since the original images in the original image set do not necessarily have tumors, a plurality of standard images containing the tumors can be output by drawing the target object position credibility curve, so that the detection precision is improved.
In the embodiment of the invention, the confidence degree analysis is carried out on the original image set through the target object positioning confidence degree, and the standard image is obtained through the confidence degree curve obtained through the confidence degree analysis, so that the accuracy of target object detection can be improved.
The target object primary detection module 103 is configured to perform target object primary detection on the standard image by using a pre-constructed image detection network, obtain a target object primary detection result, and calculate a primary detection score of the target object based on the target object primary detection result.
In the embodiment of the present invention, the pre-constructed image detection network may be a fast RCNN detection model, where the fast RCNN detection model includes a conversion layer (Conv filters), a RPN (Region Proposal Networks) network, a Pooling layer (Roi Pooling), and a Classification layer (Classification). The transformation layer (Conv layers) transforms the standard image into a feature map by convolution, which is a linear operation, and the convolution operation of the image can eliminate noise and enhance features. The RPN network comprises anchor frame generation, detection function judgment and frame regression, a candidate region is obtained by generating a real boundary frame and a series of anchor frames in the feature map, whether a target exists in the candidate region is judged by using a detection function, and frame regression is carried out on the candidate region with the target, so that an accurate candidate region (namely a region in the candidate frame) is obtained. The anchor frame is a prediction boundary frame generated by taking pixel points as centers and collecting a large number of areas in the characteristic map. The detection function may use a softmax function. The frame regression refers to a process of approaching the generated prediction boundary frame with the marked real boundary frame as a target in the target detection process. The exact candidate regions may vary in size, and the exact candidate region is converted to a fixed-size image using the Pooling layer (Roi Pooling). The Classification layer (Classification) is used to determine which category (e.g., whether a tumor) the target in the precise candidate region specifically belongs to and to output a primary score for each primary layer.
Specifically, the target object primary detection module 103 obtains the primary detection score of the standard image by:
performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, wherein the primary detection result of the target object comprises a candidate frame and confidence corresponding to the candidate frame;
and carrying out normalization processing on the candidate frames, and obtaining the primary detection score of the standard image based on the confidence coefficient corresponding to the candidate frames.
In the embodiment of the present invention, the target object primary detection module 103 obtains the primary detection score of the standard image by the following operations:
the preliminary score was calculated using the following formula:
wherein ,
for the primary check score of the mth standard image, < >>
Confidence of kth candidate frame for mth standard image, a m,k A value in the interval 0 to 1, +.>
The number of candidate frames in the mth standard image.
In the embodiment of the present invention, the normalization processing may be that the mth standard image, the kth candidate frame area is divided by all the candidate frame areas in the standard image. And the standard image is finely positioned by utilizing the pre-constructed image detection network, so that the accuracy of target object detection is further improved.
The image selection module 104 is configured to select a candidate image from the standard images according to the initial detection score and the target object positioning confidence.
Specifically, the image selection module 104 selects a candidate image from the standard images by:
adding and sequencing the initial detection scores of the standard images and the target object positioning confidence corresponding to the standard images;
and selecting a preset number of standard images from high to low as candidate images based on the added scores. Wherein the detection score of each standard image is calculated by
And the corresponding target positioning confidence level->
Adding, wherein a preset number of detection layers is selected from the high score to the low score, and the preset number can be 40% of the number of the main layers.
The target object final inspection module 105 is configured to perform target object final inspection on the candidate image by using the image detection network to obtain a target object final inspection result, and screen a final target object from the target object final inspection result by using a preset voting mechanism.
In the embodiment of the invention, the pre-constructed image detection network may also be a fast RCNN detection model, and the final detection result of the target object includes a detection frame and a confidence coefficient corresponding to the detection frame.
Specifically, the target final inspection module 105 screens out a final target from the target final inspection result by:
the target object final detection result comprises detection frames and confidence degrees corresponding to the detection frames, pixel points with the highest confidence degrees in each detection frame are selected, and confidence average calculation is carried out on all detection frames containing the pixel points to obtain final detection scores;
and selecting the target object in the detection frame with the highest final detection score as the final target object.
In the embodiment of the invention, two pixel points R are used 1 and R2 For example, assume that there are two detection boxes (BBOX) for each pixel, where R 1 The corresponding detection frames are respectively
The corresponding confidence is
The corresponding confidence is->
R 2 The corresponding detection frames are respectively
The corresponding confidence is->
The corresponding confidence is->
The final detection scores are +.>
And selecting the target object in the detection frame with the highest final detection score as the final detection result.
In the embodiment of the invention, the final detection result of the target object is optimized based on the voting mechanism, so that the final target object is obtained, and the target object detection efficiency is improved.
Fig. 7 is a schematic structural diagram of an electronic device for implementing a target object detection method according to an embodiment of the present invention.
The electronic device 1 may comprise a processor 10, a memory 11 and a bus, and may further comprise a computer program, such as an object detection program 12, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, such as a removable hard disk of the electronic device 1. The memory 11 may in other embodiments also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device 1. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 11 may be used not only for storing application software installed in the electronic device 1 and various types of data, such as codes of the object detection program 12, but also for temporarily storing data that has been output or is to be output.
The processor 10 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects respective parts of the entire electronic device using various interfaces and lines, and executes various functions of the electronic device 1 and processes data by running or executing programs or modules (e.g., object detection programs, etc.) stored in the memory 11, and calling data stored in the memory 11.
The bus may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
Fig. 7 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 7 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device 1 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.
Further, the electronic device 1 may also comprise a network interface, optionally the network interface may comprise a wired interface and/or a wireless interface (e.g. WI-FI interface, bluetooth interface, etc.), typically used for establishing a communication connection between the electronic device 1 and other electronic devices.
The electronic device 1 may optionally further comprise a user interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device 1 and for displaying a visual user interface.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The object detection program 12 stored in the memory 11 of the electronic device 1 is a combination of a plurality of instructions, which when executed in the processor 10, can implement:
performing coarse positioning of a target object on an original image set by using a pre-constructed image classification network, and obtaining the positioning confidence coefficient of the target object of each image in the original image set;
performing confidence analysis on the original image set according to the target object positioning confidence, and selecting a plurality of images from the original image set to serve as standard images based on the confidence analysis;
Performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, and calculating the primary detection score of the target object based on the primary detection result of the target object;
selecting candidate images from the standard images according to the initial detection score and the target object positioning confidence;
and carrying out target object final inspection on the candidate images by utilizing the image detection network to obtain a target object final inspection result, and screening out a final target object from the target object final inspection result by utilizing a preset voting mechanism.
Specifically, the specific implementation method of the above instructions by the processor 10 may refer to descriptions of related steps in the corresponding embodiments of fig. 1 to 5, which are not repeated herein.
Further, the modules/units integrated in the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. The computer readable storage medium may be volatile or nonvolatile. For example, the computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
The present invention also provides a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, can implement:
performing coarse positioning of a target object on an original image set by using a pre-constructed image classification network, and obtaining the positioning confidence coefficient of the target object of each image in the original image set;
performing confidence analysis on the original image set according to the target object positioning confidence, and selecting a plurality of images from the original image set to serve as standard images based on the confidence analysis;
performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, and calculating the primary detection score of the target object based on the primary detection result of the target object;
selecting candidate images from the standard images according to the initial detection score and the target object positioning confidence;
and carrying out target object final inspection on the candidate images by utilizing the image detection network to obtain a target object final inspection result, and screening out a final target object from the target object final inspection result by utilizing a preset voting mechanism.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (7)
1. A method of detecting a target, the method comprising:
performing coarse positioning of a target object on an original image set by using a pre-constructed image classification network, and obtaining the positioning confidence coefficient of the target object of each image in the original image set;
performing confidence analysis on the original image set according to the target object positioning confidence, and selecting a plurality of images from the original image set to serve as standard images based on the confidence analysis;
performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, and calculating the primary detection score of the target object based on the primary detection result of the target object;
selecting candidate images from the standard images according to the initial detection score and the target object positioning confidence;
performing target object final inspection on the candidate images by using the image detection network to obtain target object final inspection results, and screening out final target objects from the target object final inspection results by using a preset voting mechanism;
the confidence analysis is performed on the original image set according to the object positioning confidence, and a plurality of images are selected from the original image set to serve as standard images based on the confidence analysis, and the method comprises the following steps: constructing a coordinate axis by utilizing the layer direction of the image layer and the target object positioning confidence, and drawing a positioning confidence curve in the coordinate axis; calculating a half-width value corresponding to each peak value in the position credibility curve; taking an original image layer where the half-width value is located as a starting layer and an ending layer, and acquiring an image layer between the starting layer and the ending layer as the standard image;
The method for performing the primary detection of the target object on the standard image by utilizing the pre-constructed image detection network to obtain a primary detection result of the target object, and calculating the primary detection score of the target object based on the primary detection result of the target object comprises the following steps: performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, wherein the primary detection result of the target object comprises a candidate frame and confidence corresponding to the candidate frame; normalizing the candidate frames, and obtaining the primary detection score of the standard image based on the confidence coefficient corresponding to the candidate frames;
the normalizing processing is carried out on the candidate frames, and the initial detection score of the standard image is obtained based on the confidence coefficient corresponding to the candidate frames, and the normalizing processing comprises the following steps: the preliminary score was calculated using the following formula:
wherein ,
for the primary check score of the mth standard image, < >>
Confidence of kth candidate frame for mth standard image, a m,k A value in the interval 0 to 1, +.>
The number of candidate frames in the mth standard image.
2. The method for detecting an object according to claim 1, wherein the performing object coarse positioning on an original image set by using a pre-constructed image classification network, and obtaining the object positioning confidence of each image in the original image set, includes:
Layering the obtained target object image to obtain an original image set consisting of a plurality of image layers;
and performing classification prediction on each image in the original image set layer by utilizing the pre-constructed image classification network to obtain the target object positioning confidence of each image.
3. The method of claim 1, wherein selecting a candidate image from the standard image based on the preliminary detection score and the target positioning confidence comprises:
adding and sequencing the initial detection scores of the standard images and the target object positioning confidence corresponding to the standard images;
and selecting a preset number of standard images from high to low as candidate images based on the added scores.
4. The method for detecting a target object according to claim 3, wherein the step of screening the final target object from the final target object detection result by using a preset voting mechanism comprises:
the target object final detection result comprises detection frames and confidence degrees corresponding to the detection frames, pixel points with the highest confidence degrees in each detection frame are selected, and confidence average calculation is carried out on all detection frames containing the pixel points to obtain final detection scores;
And selecting the target object in the detection frame with the highest final detection score as the final target object.
5. An object detection device, the device comprising:
the target object coarse positioning module is used for performing target object coarse positioning on an original image set by utilizing a pre-constructed image classification network, and obtaining the target object positioning confidence coefficient of each image in the original image set;
the confidence analysis module is used for carrying out confidence analysis on the original image set according to the object positioning confidence, and selecting a plurality of images from the original image set to serve as standard images based on the confidence analysis;
the target object primary detection module is used for carrying out target object primary detection on the standard image by utilizing a pre-constructed image detection network to obtain a target object primary detection result, and calculating the primary detection score of the target object based on the target object primary detection result;
the image selecting module is used for selecting candidate images from the standard images according to the initial detection score and the target object positioning confidence;
the target object final inspection module is used for performing target object final inspection on the candidate images by utilizing the image detection network to obtain target object final inspection results, and screening out final target objects from the target object final inspection results by utilizing a preset voting mechanism;
The confidence analysis is performed on the original image set according to the object positioning confidence, and a plurality of images are selected from the original image set to serve as standard images based on the confidence analysis, and the method comprises the following steps: constructing a coordinate axis by utilizing the layer direction of the image layer and the target object positioning confidence, and drawing a positioning confidence curve in the coordinate axis; calculating a half-width value corresponding to each peak value in the position credibility curve; taking an original image layer where the half-width value is located as a starting layer and an ending layer, and acquiring an image layer between the starting layer and the ending layer as the standard image;
the method for performing the primary detection of the target object on the standard image by utilizing the pre-constructed image detection network to obtain a primary detection result of the target object, and calculating the primary detection score of the target object based on the primary detection result of the target object comprises the following steps: performing primary detection on the standard image by using a pre-constructed image detection network to obtain a primary detection result of the target object, wherein the primary detection result of the target object comprises a candidate frame and confidence corresponding to the candidate frame; normalizing the candidate frames, and obtaining the primary detection score of the standard image based on the confidence coefficient corresponding to the candidate frames;
The normalizing processing is carried out on the candidate frames, and the initial detection score of the standard image is obtained based on the confidence coefficient corresponding to the candidate frames, and the normalizing processing comprises the following steps: the preliminary score was calculated using the following formula:
wherein ,
for the primary check score of the mth standard image, < >>
Confidence of kth candidate frame for mth standard image, a m,k A value in the interval 0 to 1, +.>
The number of candidate frames in the mth standard image.
6. An electronic device, the electronic device comprising:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the object detection method according to any one of claims 1 to 4.
7. A computer-readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the object detection method according to any one of claims 1 to 4.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110325008.5A CN113160144B (en) | 2021-03-25 | 2021-03-25 | Target object detection method, target object detection device, electronic equipment and storage medium |
| PCT/CN2021/097076 WO2022198786A1 (en) | 2021-03-25 | 2021-05-30 | Target object detection method and apparatus, and electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110325008.5A CN113160144B (en) | 2021-03-25 | 2021-03-25 | Target object detection method, target object detection device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113160144A CN113160144A (en) | 2021-07-23 |
| CN113160144B true CN113160144B (en) | 2023-05-26 |
Family
ID=76884885
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110325008.5A Active CN113160144B (en) | 2021-03-25 | 2021-03-25 | Target object detection method, target object detection device, electronic equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113160144B (en) |
| WO (1) | WO2022198786A1 (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019100946A1 (en) * | 2017-11-22 | 2019-05-31 | 北京市商汤科技开发有限公司 | Object detection method, device, and apparatus |
| CN110084173A (en) * | 2019-04-23 | 2019-08-02 | 精伦电子股份有限公司 | Number of people detection method and device |
| CN111275082A (en) * | 2020-01-14 | 2020-06-12 | 中国地质大学（武汉） | Indoor object target detection method based on improved end-to-end neural network |
| CN111553947A (en) * | 2020-04-17 | 2020-08-18 | 腾讯科技（深圳）有限公司 | Target object positioning method and device |
| CN111932482A (en) * | 2020-09-25 | 2020-11-13 | 平安科技（深圳）有限公司 | Method and device for detecting target object in image, electronic equipment and storage medium |
| WO2021017998A1 (en) * | 2019-07-26 | 2021-02-04 | 第四范式（北京）技术有限公司 | Method and system for positioning text position, and method and system for training model |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3432784B1 (en) * | 2016-03-24 | 2020-09-23 | The Regents of The University of California | Deep-learning-based cancer classification using a hierarchical classification framework |
| CN111339891A (en) * | 2020-02-20 | 2020-06-26 | 苏州浪潮智能科技有限公司 | Target detection method of image data and related device |
| CN111583204B (en) * | 2020-04-27 | 2022-10-14 | 天津大学 | Organ positioning method of two-dimensional sequence magnetic resonance image based on network model |
| CN112465060A (en) * | 2020-12-10 | 2021-03-09 | 平安科技（深圳）有限公司 | Method and device for detecting target object in image, electronic equipment and readable storage medium |
-
2021
- 2021-03-25 CN CN202110325008.5A patent/CN113160144B/en active Active
- 2021-05-30 WO PCT/CN2021/097076 patent/WO2022198786A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019100946A1 (en) * | 2017-11-22 | 2019-05-31 | 北京市商汤科技开发有限公司 | Object detection method, device, and apparatus |
| CN110084173A (en) * | 2019-04-23 | 2019-08-02 | 精伦电子股份有限公司 | Number of people detection method and device |
| WO2021017998A1 (en) * | 2019-07-26 | 2021-02-04 | 第四范式（北京）技术有限公司 | Method and system for positioning text position, and method and system for training model |
| CN111275082A (en) * | 2020-01-14 | 2020-06-12 | 中国地质大学（武汉） | Indoor object target detection method based on improved end-to-end neural network |
| CN111553947A (en) * | 2020-04-17 | 2020-08-18 | 腾讯科技（深圳）有限公司 | Target object positioning method and device |
| CN111932482A (en) * | 2020-09-25 | 2020-11-13 | 平安科技（深圳）有限公司 | Method and device for detecting target object in image, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113160144A (en) | 2021-07-23 |
| WO2022198786A1 (en) | 2022-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111932482B (en) | Method and device for detecting target object in image, electronic equipment and storage medium | |
| CN112395978B (en) | Behavior detection method, behavior detection device and computer readable storage medium | |
| CN113283446B (en) | Method and device for identifying object in image, electronic equipment and storage medium | |
| CN111915609B (en) | Focus detection analysis method, apparatus, electronic device and computer storage medium | |
| CN111932562B (en) | Image identification method and device based on CT sequence, electronic equipment and medium | |
| CN111932547B (en) | Method and device for segmenting target object in image, electronic device and storage medium | |
| CN111695609B (en) | Target damage degree judging method and device, electronic equipment and storage medium | |
| CN111933274B (en) | Disease classification diagnosis method, device, electronic equipment and storage medium | |
| CN111696084B (en) | Cell image segmentation method, device, electronic equipment and readable storage medium | |
| CN111783982B (en) | Method, device, equipment and medium for acquiring attack sample | |
| CN111862096B (en) | Image segmentation method and device, electronic equipment and storage medium | |
| CN111932534A (en) | Medical image picture analysis method and device, electronic equipment and readable storage medium | |
| CN111476760B (en) | Medical image generation method and device, electronic equipment and medium | |
| CN113516417A (en) | Service evaluation method and device based on intelligent modeling, electronic equipment and medium | |
| CN112132812B (en) | Certificate verification method and device, electronic equipment and medium | |
| CN112990374A (en) | Image classification method, device, electronic equipment and medium | |
| CN113487621A (en) | Medical image grading method and device, electronic equipment and readable storage medium | |
| CN116741358A (en) | Inquiry registration recommendation method, inquiry registration recommendation device, inquiry registration recommendation equipment and storage medium | |
| CN113160144B (en) | Target object detection method, target object detection device, electronic equipment and storage medium | |
| CN115294426B (en) | Method, device and equipment for tracking interventional medical equipment and storage medium | |
| CN114511569B (en) | Tumor marker-based medical image identification method, device, equipment and medium | |
| CN113627394B (en) | Face extraction method and device, electronic equipment and readable storage medium | |
| CN113705686B (en) | Image classification method, device, electronic equipment and readable storage medium | |
| CN112580505B (en) | Method and device for identifying network point switch door state, electronic equipment and storage medium | |
| CN112233194B (en) | Medical picture optimization method, device, equipment and computer readable 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 |