CN111899264A - Target image segmentation method, device and medium - Google Patents
Target image segmentation method, device and medium Download PDFInfo
- Publication number
- CN111899264A CN111899264A CN202010561367.6A CN202010561367A CN111899264A CN 111899264 A CN111899264 A CN 111899264A CN 202010561367 A CN202010561367 A CN 202010561367A CN 111899264 A CN111899264 A CN 111899264A
- Authority
- CN
- China
- Prior art keywords
- target image
- segmentation
- grade
- target
- image
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000003709 image segmentation Methods 0.000 title claims abstract description 37
- 230000011218 segmentation Effects 0.000 claims abstract description 118
- 238000013528 artificial neural network Methods 0.000 claims abstract description 36
- 238000012549 training Methods 0.000 claims abstract description 17
- 230000003321 amplification Effects 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 abstract description 17
- 238000012545 processing Methods 0.000 abstract description 10
- 230000006870 function Effects 0.000 description 9
- 230000006872 improvement Effects 0.000 description 9
- 238000003860 storage Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000005520 cutting process Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000003068 static 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
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a target image segmentation method, device and medium, comprising: inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples; and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks. In the embodiment of the description, the grade of the target image is determined, and then the segmentation processing is performed according to the segmentation network corresponding to the grade of the target image, so that the segmentation effect of the segmentation network is better.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, and a medium for segmenting a target image.
Background
The target segmentation is to segment the target in the picture. For example, if the set division target is a person, a new picture needs to be finally divided from the picture, and only the set target person but nothing else should be in the divided picture. Target segmentation is an important branch in the field of computer vision, and has wide application in the fields of scene understanding, lesion segmentation, automatic driving and the like. The existing target segmentation technology has poor effect on segmenting images and cannot segment targets well.
Disclosure of Invention
In view of this, embodiments of the present application provide a method, an apparatus, and a medium for segmenting a target image, which are used to solve the problem in the prior art that the target segmentation technique is poor in effect when segmenting an image.
The embodiment of the application adopts the following technical scheme:
the embodiment of the application provides a target image segmentation method, which comprises the following steps:
inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples;
and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
It should be noted that, in the embodiments of the present specification, the level of the target image is determined, and then the segmentation processing is performed according to the segmentation network corresponding to the level of the target image, so that the segmentation effect of the segmentation network is better.
Further, the levels of the target images include a first level and a second level, wherein the resolution of the target image corresponding to the first level is greater than the resolution of the target image corresponding to the second level.
It should be noted that, in the prior art, for a target image with a low resolution, the effect is not good when the cutting network is directly cut, and in the embodiment of the present specification, the target image with the low resolution can be distinguished by dividing the target image into two levels, so that the final cutting effect is better.
Further, the split networks include a first split network corresponding to the first level and a second split network corresponding to the second level, where the first split network includes a full convolutional neural network module, and the second split network includes an amplifying module and a full convolutional neural network module.
It should be noted that, in the embodiments of the present specification, different segmentation networks may be adopted for target images with different resolutions, so that the final cutting effect is better.
Further, the sending the target image to a segmentation network corresponding to the target image level to determine a segmentation result of the target image specifically includes:
if the grade of the target image is the first grade, the target image is sent to the first segmentation network, image segmentation is carried out on the target image through a full convolution neural network module of the first segmentation network, and a first segmentation result of the target image is determined;
and if the grade of the target image is the second grade, sending the target image to the second segmentation network, amplifying the target image through an amplification module of the second segmentation network, and performing image segmentation on the amplified target image through a full convolution neural network module of the second segmentation network to determine a second segmentation result of the target image.
It should be noted that, in the embodiments of the present description, different segmentation networks may be adopted for target images with different resolutions, and if the target image is a first-level target image with a high resolution, the full convolution neural network module is directly adopted to perform image segmentation; if the target image is of a second level with lower resolution, the target image can be amplified through the amplifying module, and then the full convolution neural network module is adopted to perform image segmentation, so that the final cutting effect is better.
Further, the target level recognition network is a neural network trained based on a residual network model structure.
It should be noted that, after the residual error network model structure is adopted in the embodiment of the present application, a phenomenon that an error increases on a training set due to a continuously deepened number of layers is eliminated, a training error of the residual error network model gradually decreases along with the increase of the number of layers, and performance on a test set also becomes better, so that a final training of a target level recognition network is more accurate.
Further, the full convolution neural network module is a neural network module trained based on an SE module; the amplification module is a neural network module trained based on a BEGAN module.
It should be noted that, the SE module adopted in the embodiment of the present application can adaptively select the feature map that is helpful for improving the segmentation performance, and therefore, placing the SE module behind the pooling layer of the full convolution neural network can improve the distinctiveness of the features and improve the segmentation performance of the entire segmentation network. The neural network module trained by the BEGAN module can increase the resolution of a target image, and then the neural network module trained on the SE module is used for segmentation, so that the distinguishing performance of features can be improved, and the segmentation performance of the whole segmentation network can be improved.
Further, the method further comprises:
inputting the traffic driving image into a pre-trained target grade recognition network, and determining the grade of the traffic driving image;
and sending the traffic driving image to a segmentation network corresponding to the grade of the traffic driving image, and determining a segmentation result of the traffic driving image.
It should be noted that, in the embodiment of the present application, the target image segmentation system is applied in a traffic driving scene, and a segmentation result in a traffic driving image can be determined.
Further, the traffic driving image comprises one or more of a traffic light image, a speed limit sign image and a road indication image.
It should be noted that after the traffic light image, the speed limit identification image and the road indication image are separated from the traffic driving image, relevant information can be provided for the driver, and the driving safety is ensured.
An embodiment of the present application further provides a target image segmentation apparatus, where the apparatus includes:
at least one processor; and the number of the first and second groups,
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:
inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples;
and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
An embodiment of the present application further provides a target image segmentation medium, in which computer-executable instructions are stored, where the computer-executable instructions are configured to:
inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples;
and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
The embodiment of the application adopts at least one technical scheme which can achieve the following beneficial effects: in the embodiment of the description, the grade of the target image is determined, and then the segmentation processing is performed according to the segmentation network corresponding to the grade of the target image, so that the segmentation effect of the segmentation network is better.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic flowchart of a target image segmentation method provided in an embodiment of the present disclosure;
fig. 2 is a schematic flowchart of a target image segmentation method provided in the second embodiment of the present specification.
Detailed Description
The existing target segmentation technology can achieve a satisfactory effect to a certain extent, but the effect is not good when segmentation is performed on certain targets with low resolution, and effective segmentation cannot be achieved.
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic flowchart of a target image segmentation method provided in an embodiment of the present specification, where the embodiment of the present specification may be implemented by a target image segmentation system, and the method specifically includes:
step S101, inputting a target image into a pre-trained target grade recognition network by a target image segmentation system, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples.
In step S101 in the embodiment of the present specification, in the related art, when a target image with low resolution is divided, the effect is not good. In the embodiment of the specification, the target images are classified, so that the target image segmentation system can achieve a good effect in processing target images of different grades.
Step S102, the target image segmentation system sends the target image to segmentation networks corresponding to the target image grades, and determines the segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
Corresponding to the first embodiment of the present specification, fig. 2 is a schematic flowchart of a target image segmentation method provided by the second embodiment of the present specification, and the following steps may be executed by the target image segmentation system in the embodiment of the present specification, which specifically include:
in step S201, the target image segmentation system constructs a data set.
In step S201 of the embodiment of the present specification, a plurality of samples of the mark target image level are included in the data set. The levels of the target images include a first level and a second level, wherein the images of the first level may represent the target images with high resolution, and the images of the second level may represent the images with low resolution. The condition for determining the rank of the target image may be set in advance, for example, a first rank (high resolution) in which the resolution is higher than 800 × 600 and a second rank (low resolution) in which the resolution is not higher than 800 × 600 may be set. In the prior art, when a target image of a second level is segmented, the effect is not good, and in the embodiment of the specification, the target image segmentation system can achieve a good effect in processing target images of different levels by classifying the target image.
In step S202, the target image segmentation system establishes an initial target level identification network.
In step S202 of the embodiment of the present specification, the target level recognition network may be a neural network trained based on a residual network model structure. Wherein, the residual network model is the ResNet network model structure. After the ResNet network model structure is used, the phenomenon that errors on a training set are increased due to the fact that the number of layers is continuously deepened is eliminated, the training errors of the ResNet network model can be gradually reduced along with the increase of the number of layers, performance on a test set is good, and the final training of the target grade recognition network is more accurate.
Step S203, training the initial target level identification network according to the data set to obtain the target level identification network meeting the requirements.
In step S203 of the embodiment of the present specification, the obtained target level recognition network may perform level recognition on the unknown target image, so as to classify whether the target image is at the first level or the second level.
And step S204, inputting the target image into a target grade recognition network by the target image segmentation system, and determining the grade of the target image.
In step S204 of the embodiment of the present specification, the levels of the target images include a first level and a second level, wherein the resolution of the target images corresponding to the first level may be greater than the resolution of the target images corresponding to the second level.
It should be noted that the levels of the target image may further include a first level, a second level, and a third level, and the specific situation may be determined according to the actual situation, which is not limited in the present application, and the larger the number of the division levels is, the more the division networks corresponding to different levels are, and the better the final division effect is.
Step S205, the target image segmentation system sends the target image to the segmentation networks corresponding to the target image grades, and determines the segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
In step S205 of this embodiment, the split networks may include a first split network corresponding to the first level and a second split network corresponding to the second level, where the first split network includes a full convolutional neural network module, and the second split network includes an amplifying module and a full convolutional neural network module.
Sending the target image to a segmentation network corresponding to the target image level, and determining a segmentation result of the target image, which specifically comprises the following steps:
if the grade of the target image is the first grade, the target image is sent to the first segmentation network, image segmentation is carried out on the target image through a full convolution neural network module of the first segmentation network, and a first segmentation result of the target image is determined;
and if the grade of the target image is the second grade, sending the target image to the second segmentation network, amplifying the target image through an amplification module of the second segmentation network, and performing image segmentation on the amplified target image through a full convolution neural network module of the second segmentation network to determine a second segmentation result of the target image.
Wherein, the full convolution neural network module can be a neural network module trained based on an SE (Squeeze-and-Excitation) module; the amplification module may be a neural network module trained based on the BEGAN module.
The SE module can adaptively select the feature map which is beneficial to improving the segmentation performance, so that the feature distinguishability can be improved and the segmentation performance of the whole segmentation network can be improved by placing the SE module behind the pooling layer of the full convolution neural network.
If the level of the target image is the second level, the effect is not good if the target image is directly segmented due to low resolution, and a neural network module trained by the BEGAN module can be used for generating a high-resolution image. In the image generated by the BEGAN module, the resolution of the target image can be increased, and then the neural network module trained based on the SE module is used for segmentation, so that the distinguishing performance of the features can be improved, and the segmentation performance of the whole segmentation network can be improved.
It should be noted that the pixels of the target image corresponding to the first level may be larger than the pixels of the target image corresponding to the second level, that is, the target image may also be classified according to the size of the pixel values. Accordingly, the pixel values of the target image may also be increased in the image generated by the BEGAN module.
It should be noted that the object recognition is a classification task, i.e., it is recognized from the image which are the first type of object, which are the second type of object, and which are the third type of object, and there is a substantial difference from the object segmentation.
Further, in an embodiment of the present specification, if the target image segmentation system is applied to the aspect of traffic, the method may specifically include:
inputting the traffic driving image into a pre-trained target grade recognition network to determine the grade of the traffic driving image;
and sending the traffic driving image to a segmentation network corresponding to the grade of the traffic driving image, and determining a segmentation result of the traffic driving image.
The traffic driving image comprises one or more of a traffic light image, a speed limit identification image and a road indication image.
An embodiment of the present application further provides a target image segmentation apparatus, where the apparatus includes:
at least one processor; and the number of the first and second groups,
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:
inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples;
and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
An embodiment of the present application further provides a target image segmentation medium, in which computer-executable instructions are stored, where the computer-executable instructions are configured to:
inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples;
and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Language Description Language), traffic, pl (core unified Programming Language), HDCal, JHDL (Java Hardware Description Language), langue, Lola, HDL, laspam, hardsradware (Hardware Description Language), vhjhd (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.
The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method of segmenting a target image, the method comprising:
inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples;
and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
2. The method of claim 1, wherein the target image levels comprise a first level and a second level, and wherein the resolution of the target image corresponding to the first level is greater than the resolution of the target image corresponding to the second level.
3. The method of claim 2, wherein the segmentation network comprises a first segmentation network corresponding to the first level and a second segmentation network corresponding to the second level, wherein the first segmentation network comprises a full convolutional neural network module, and the second segmentation network comprises an amplification module and a full convolutional neural network module.
4. The method for segmenting the target image according to claim 3, wherein the step of sending the target image to the segmentation network corresponding to the grade of the target image to determine the segmentation result of the target image specifically comprises the steps of:
if the grade of the target image is the first grade, the target image is sent to the first segmentation network, image segmentation is carried out on the target image through a full convolution neural network module of the first segmentation network, and a first segmentation result of the target image is determined;
and if the grade of the target image is the second grade, sending the target image to the second segmentation network, amplifying the target image through an amplification module of the second segmentation network, and performing image segmentation on the amplified target image through a full convolution neural network module of the second segmentation network to determine a second segmentation result of the target image.
5. The target image segmentation method of claim 1 wherein the target level recognition network is a neural network trained based on a residual network model structure.
6. The target image segmentation method of claim 3 wherein the fully convolutional neural network module is a neural network module trained based on an SE module; the amplification module is a neural network module trained based on a BEGAN module.
7. The method of target image segmentation according to claim 1, further comprising:
inputting the traffic driving image into a pre-trained target grade recognition network, and determining the grade of the traffic driving image;
and sending the traffic driving image to a segmentation network corresponding to the grade of the traffic driving image, and determining a segmentation result of the traffic driving image.
8. The object image segmentation method according to claim 7, wherein the traffic driving image includes one or more of a traffic light image, a speed limit sign image, and a road indication image.
9. An apparatus for segmenting an object image, the apparatus comprising:
at least one processor; and the number of the first and second groups,
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:
inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples;
and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
10. A target image segmentation medium storing computer-executable instructions, the computer-executable instructions configured to:
inputting a target image into a pre-trained target grade recognition network, and determining the grade of the target image, wherein the target grade recognition network is obtained by training according to a plurality of data sets marking target image grade samples;
and sending the target image to a segmentation network corresponding to the target image grade, and determining a segmentation result of the target image, wherein different target grades correspond to different segmentation networks, and the segmentation networks are pre-trained full convolution neural networks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010561367.6A CN111899264A (en) | 2020-06-18 | 2020-06-18 | Target image segmentation method, device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010561367.6A CN111899264A (en) | 2020-06-18 | 2020-06-18 | Target image segmentation method, device and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111899264A true CN111899264A (en) | 2020-11-06 |
Family
ID=73207407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010561367.6A Pending CN111899264A (en) | 2020-06-18 | 2020-06-18 | Target image segmentation method, device and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111899264A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113673601A (en) * | 2021-08-23 | 2021-11-19 | 北京三快在线科技有限公司 | Behavior recognition method and device, storage medium and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107909053A (en) * | 2017-11-30 | 2018-04-13 | 济南浪潮高新科技投资发展有限公司 | A kind of method for detecting human face based on grade study concatenated convolutional neutral net |
| CN108596115A (en) * | 2018-04-27 | 2018-09-28 | 济南浪潮高新科技投资发展有限公司 | A kind of vehicle checking method, apparatus and system based on convolutional neural networks |
| CN109214388A (en) * | 2018-11-21 | 2019-01-15 | 济南浪潮高新科技投资发展有限公司 | A kind of lesion segmentation approach and device based on personalized converged network |
| US20190080456A1 (en) * | 2017-09-12 | 2019-03-14 | Shenzhen Keya Medical Technology Corporation | Method and system for performing segmentation of image having a sparsely distributed object |
| CN109528230A (en) * | 2018-11-21 | 2019-03-29 | 济南浪潮高新科技投资发展有限公司 | A kind of tumor of breast dividing method and device based on multi-stage transformation network |
| CN110929788A (en) * | 2019-11-22 | 2020-03-27 | 珠海格力电器股份有限公司 | Food material identification method and device, storage medium and electronic equipment |
-
2020
- 2020-06-18 CN CN202010561367.6A patent/CN111899264A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190080456A1 (en) * | 2017-09-12 | 2019-03-14 | Shenzhen Keya Medical Technology Corporation | Method and system for performing segmentation of image having a sparsely distributed object |
| CN107909053A (en) * | 2017-11-30 | 2018-04-13 | 济南浪潮高新科技投资发展有限公司 | A kind of method for detecting human face based on grade study concatenated convolutional neutral net |
| CN108596115A (en) * | 2018-04-27 | 2018-09-28 | 济南浪潮高新科技投资发展有限公司 | A kind of vehicle checking method, apparatus and system based on convolutional neural networks |
| CN109214388A (en) * | 2018-11-21 | 2019-01-15 | 济南浪潮高新科技投资发展有限公司 | A kind of lesion segmentation approach and device based on personalized converged network |
| CN109528230A (en) * | 2018-11-21 | 2019-03-29 | 济南浪潮高新科技投资发展有限公司 | A kind of tumor of breast dividing method and device based on multi-stage transformation network |
| CN110929788A (en) * | 2019-11-22 | 2020-03-27 | 珠海格力电器股份有限公司 | Food material identification method and device, storage medium and electronic equipment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113673601A (en) * | 2021-08-23 | 2021-11-19 | 北京三快在线科技有限公司 | Behavior recognition method and device, storage medium and electronic equipment |
| CN113673601B (en) * | 2021-08-23 | 2023-02-03 | 北京三快在线科技有限公司 | Behavior recognition method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111311709B (en) | Method and device for generating high-precision map | |
| CN109034183B (en) | Target detection method, device and equipment | |
| CN111508258B (en) | Positioning method and device | |
| CN111639682A (en) | Ground segmentation method and device based on point cloud data | |
| CN112732585A (en) | Simulation test method and device | |
| CN113887608B (en) | Model training method, image detection method and device | |
| CN112308113A (en) | Target identification method, device and medium based on semi-supervision | |
| CN114440903A (en) | High-precision map construction method and device, storage medium and electronic equipment | |
| CN112766241A (en) | Target object identification method and device | |
| CN114332873A (en) | Training method and device for recognition model | |
| CN115828162A (en) | Classification model training method and device, storage medium and electronic equipment | |
| CN117197781B (en) | Traffic sign recognition method and device, storage medium and electronic equipment | |
| CN114626437A (en) | Model training method and device, storage medium and electronic equipment | |
| CN111899264A (en) | Target image segmentation method, device and medium | |
| CN112990099A (en) | Method and device for detecting lane line | |
| CN112699043A (en) | Method and device for generating test case | |
| CN116805393A (en) | Hyperspectral image classification method and system based on 3DUnet spectrum-space information fusion | |
| CN111539962A (en) | Target image classification method, device and medium | |
| CN111539961A (en) | Target segmentation method, device and equipment | |
| CN116246276A (en) | Information identification method, device, equipment and readable storage medium | |
| CN113344198B (en) | Model training method and device | |
| CN114359935A (en) | Model training and form recognition method and device | |
| CN112560530B (en) | Two-dimensional code processing method, device, medium and electronic device | |
| CN114187355A (en) | Image calibration method and device | |
| CN113706552A (en) | Method and device for generating semantic segmentation marking data of laser reflectivity base map |
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 |