WO2021036382A1 - 图像处理方法及装置、电子设备和存储介质 - Google Patents

图像处理方法及装置、电子设备和存储介质 Download PDF

Info

Publication number
WO2021036382A1
WO2021036382A1 PCT/CN2020/093779 CN2020093779W WO2021036382A1 WO 2021036382 A1 WO2021036382 A1 WO 2021036382A1 CN 2020093779 W CN2020093779 W CN 2020093779W WO 2021036382 A1 WO2021036382 A1 WO 2021036382A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
human body
clustering
face
result
Prior art date
Application number
PCT/CN2020/093779
Other languages
English (en)
French (fr)
Other versions
WO2021036382A9 (zh
Inventor
朱铖恺
张学森
武伟
黄礼玮
梁栋
Original Assignee
深圳市商汤科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳市商汤科技有限公司 filed Critical 深圳市商汤科技有限公司
Priority to SG11202110569TA priority Critical patent/SG11202110569TA/en
Priority to JP2021552939A priority patent/JP2022523243A/ja
Publication of WO2021036382A1 publication Critical patent/WO2021036382A1/zh
Priority to US17/488,631 priority patent/US20220019772A1/en
Publication of WO2021036382A9 publication Critical patent/WO2021036382A9/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172—Classification, e.g. identification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/55—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
    • G06V10/763—Non-hierarchical techniques, e.g. based on statistics of modelling distributions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
    • G06V10/809—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of classification results, e.g. where the classifiers operate on the same input data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation

Definitions

  • the present disclosure relates to the field of artificial intelligence technology, and in particular to an image processing method and device, electronic equipment, and storage medium.
  • the image can be clustered in the manner of face clustering to establish the above-mentioned information database.
  • the present disclosure proposes a technical solution for image processing.
  • an image processing method including:
  • the image features include face features and/or human body features, and the image to be processed includes a first image and a second image ;
  • the face clustering result includes a first result, and for the first image from which the face features are extracted, the face clustering is performed according to the extracted face features Operation to obtain the face clustering results, including:
  • the face clustering result further includes a second result, and for the first image from which the face features are extracted, the face clustering is performed according to the extracted face features.
  • Class operations to obtain face clustering results including:
  • the human body clustering result includes a third result, and for the second image for which no facial features are extracted, extracting from the human face clustering result and the second image
  • the human body characteristics of and the human body characteristics in the first image from which the human body characteristics are extracted are subjected to a human body clustering operation to obtain a human body clustering result, including:
  • any of the second images perform a human body clustering operation based on the human body characteristics in the first image from which the human body characteristics are extracted and the human body characteristics in the second image to obtain a human body clustering sub-result ;
  • the second image is added to the category to which the first image belongs to the same human body category as the second image, to obtain the third result.
  • the human body clustering result further includes a fourth result, and for any second image for which no facial features are extracted, according to the face clustering result, the second Performing a human body clustering operation on the human body features extracted from the image and the human body features in the first image from which the human body features are extracted, to obtain a human body clustering result, including:
  • the method further includes:
  • the face cluster center and the human body cluster center of at least one category in the image database are updated according to the image to be processed.
  • an image processing device including:
  • the extraction module is used to perform face feature extraction and human body feature extraction on the image to be processed to obtain the image feature of the image to be processed.
  • the image feature includes the face feature and/or the human body feature, and the image to be processed includes the first Image and second image;
  • the first clustering module is configured to perform a face clustering operation according to the extracted face features for the first image from which the face features are extracted, to obtain a face clustering result;
  • the second clustering module is used for the second image for which no face features are extracted, according to the face clustering result, the human body features extracted from the second image, and the first image from which the human body features are extracted Perform the human body clustering operation on the human body features in to obtain the human body clustering result;
  • the third clustering module is configured to obtain a clustering result for the image to be processed according to the face clustering result and the human body clustering result.
  • the face clustering result includes a first result
  • the first clustering module is further configured to:
  • the face clustering result further includes a second result
  • the first clustering module is further configured to:
  • the human body clustering result includes a third result
  • the second clustering module is further used for:
  • any of the second images perform a human body clustering operation based on the human body characteristics in the first image from which the human body characteristics are extracted and the human body characteristics in the second image to obtain a human body clustering sub-result ;
  • the second image is added to the category to which the first image belongs to the same human body category as the second image, to obtain the third result.
  • the human body clustering result further includes a fourth result
  • the second clustering module is further configured to:
  • the device further includes:
  • An adding module configured to add the to-be-processed image to an image database according to the clustering result
  • the update module is used to update at least one type of face cluster center and human body cluster center in the image database according to the to-be-processed image.
  • an electronic device including:
  • a memory for storing processor executable instructions
  • the processor is configured to execute the above-mentioned image processing method.
  • a computer-readable storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, the foregoing image processing method is implemented.
  • a computer program including computer readable code, and when the computer readable code is run in an electronic device, a processor in the electronic device executes the image processing method described above. .
  • face clustering can be performed on the first image from which facial features can be extracted by face clustering, and the category to which the first image belongs can be determined to obtain the person Face clustering results.
  • Use human body clustering to cluster the second image with no facial features extracted from the first image, and then determine the category to which the second image belongs according to the face clustering result of the first image, so as to obtain the target
  • the embodiment of the present disclosure adopts a combination of face clustering and human body clustering, while ensuring the accuracy of the clustering result, the recall rate of the clustering result is improved.
  • Fig. 1 shows a flowchart of an image processing method according to an embodiment of the present disclosure
  • Fig. 2 shows a schematic diagram of an image processing method according to an embodiment of the present disclosure
  • FIG. 3 shows a schematic structural diagram of an image processing device provided by an embodiment of the present disclosure
  • Fig. 4 is a block diagram showing an electronic device 800 according to an exemplary embodiment
  • Fig. 5 is a block diagram showing an electronic device 1900 according to an exemplary embodiment.
  • Fig. 1 shows a flowchart of an image processing method according to an embodiment of the present disclosure.
  • the image processing method can be executed by a terminal device or other processing device, where the terminal device can be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, or a personal digital assistant (Personal Digital Assistant). , PDA), handheld devices, computing devices, in-vehicle devices, wearable devices, etc.
  • Other processing devices can be servers or cloud servers.
  • the image processing method may be implemented by a processor invoking computer-readable instructions stored in the memory.
  • the method includes:
  • Step S11 Perform face feature extraction and human body feature extraction on the image to be processed to obtain image features of the image to be processed.
  • the image features include face features and/or human body features, and the image to be processed includes a first image and The second image.
  • the image to be processed may be an image captured by an image capture device (such as a camera), or may be a saved image or video frame that is directly input.
  • an image capture device such as a camera
  • the image to be processed may be an image captured by an image capture device (such as a camera), or may be a saved image or video frame that is directly input.
  • a neural network such as a convolutional neural network can be used to perform facial feature extraction and human feature extraction on the image to be processed to obtain image features (face features and/or human body features) of the image to be processed, for example, for multiple images to be processed
  • image features face features and/or human body features
  • some of the images to be processed have collected facial features, and this part of the image to be processed can be determined to be the first image.
  • the first image can also be divided into An image of human body features, and an image in which both facial features and human body features are collected, and the other part of the image to be processed that has not collected facial features but collected human features can be determined as the second image.
  • the present disclosure does not limit the type of neural network, and does not limit the method of extracting facial features and human body features.
  • the face feature may be feature information determined according to key points of the face, for example, the position and shape of the facial features, and may also include information such as skin color.
  • the characteristics of the human body may be characteristic information determined according to key points of the human body, such as height, body shape, leg length, arm length, etc., and may also include information such as clothing style and color.
  • Step S12 For the first image from which the facial features are extracted, perform a face clustering operation according to the extracted facial features to obtain a face clustering result.
  • the image to be processed from which the facial features are extracted is the first image
  • the image to be processed from which the facial features are not extracted but the human features are extracted is the second image.
  • the face clustering operation can be performed based on the facial features extracted from the first image, for example: clustering multiple first images to obtain the face clustering result, or before the current clustering operation, Clustering has been performed based on historical images, and images have been stored in the image database according to existing categories.
  • the first image can be clustered into existing categories, and the first image that cannot be clustered into existing categories can be re-clustered. Perform clustering to get the face clustering result.
  • the above-mentioned face clustering operation may use any clustering method such as K-MEANS algorithm, K-MEDOIDS algorithm, CLARANS algorithm, etc.
  • the present disclosure does not specifically limit the clustering method adopted for the face clustering operation.
  • Step S13 For the second image for which no facial features are extracted, perform processing based on the face clustering result, the human body features extracted from the second image, and the human body features in the first image from which the human body features are extracted. Human body clustering operation to obtain the result of human body clustering.
  • Step S14 Obtain a clustering result for the image to be processed according to the face clustering result and the human body clustering result.
  • the human body clustering operation between the two can be performed based on the extracted human body features to determine the difference with the second image
  • the first image classified into a category, the human body clustering result is obtained.
  • the face clustering result and the human body clustering result can be merged to obtain the clustering result for the image to be processed.
  • the above-mentioned human body clustering operation may use any clustering method such as K-MEANS algorithm, K-MEDOIDS algorithm, CLARANS algorithm, etc.
  • the present disclosure does not specifically limit the clustering method adopted for the human body clustering operation.
  • face clustering can be performed on the first image from which facial features can be extracted by face clustering, and the category to which the first image belongs can be determined, and the face cluster can be obtained.
  • Class result Use human body clustering to cluster the second image with no facial features extracted from the first image, and then determine the category to which the second image belongs according to the face clustering result of the first image, so as to obtain the target
  • the embodiment of the present disclosure adopts a combination of face clustering and human body clustering, while ensuring the accuracy of the clustering result, the recall rate of the clustering result is improved.
  • the above-mentioned face clustering result may include a first result.
  • the face clustering operation is performed according to the extracted face features to obtain
  • the face clustering results can include:
  • the historical image has been clustered in advance, and the historical image has been stored in an image database according to the category determined by the clustering result, and the image database also stores the result of the clustering operation.
  • the face cluster center and the human body cluster center of at least one existing category where for any existing category, the face cluster center can be the mean value of the facial features extracted from the image corresponding to the existing category
  • the human body cluster center may be the average value of the human body features extracted from the image corresponding to the existing category.
  • At least one face cluster center of an existing category may be obtained from an image database, the similarity between the face feature of the first image and the face cluster center of at least one existing category may be determined, and then the first face cluster center may be determined according to the similarity.
  • the category to which the image belongs, the first image is clustered into existing categories, and the first result of the first image is obtained.
  • the face clustering result may include a second result.
  • the face clustering operation is performed according to the extracted face features to obtain the person
  • the face clustering results also include:
  • the face clustering operation is performed again according to the facial features extracted from the first image, and the multiple first images are clustered into The new category gets the second result of the first image.
  • first images image to be processed 1, image to be processed 2, image to be processed 3, image to be processed 4, image to be processed 5, image to be processed 6, image to be processed 7
  • image database There are 6 existing categories (person 1, person 2, person 3, person 4, person 5, person 6), and the face cluster centers of the 6 existing categories can be obtained respectively according to the 7 first images
  • the extracted face features and the face clustering centers of 6 existing categories perform the face clustering operation, and the first result is that the image to be processed 1 belongs to the person 1, the image to be processed 3 and the image 5 to be processed belong to the person 4 .
  • the face clustering operation is performed according to the extracted facial features
  • the second result is:
  • the processed image 2 and the to-be-processed image 6 belong to one category (person 7)
  • the to-be-processed image 4 belongs to one category (person 8)
  • the to-be-processed image 7 belongs to one category (person 9).
  • the first result and the second result are combined to get The result of face clustering for the first image is obtained.
  • the human body clustering result may include a third result.
  • the above-mentioned second image for which no facial features are extracted is based on the above-mentioned face clustering result.
  • any of the second images perform a human body clustering operation based on the human body characteristics in the first image from which the human body characteristics are extracted and the human body characteristics in the second image to obtain a human body clustering sub-result;
  • the second image is added to the category to which the first image belongs to the same human body category as the second image, to obtain a third result.
  • any second image it can be clustered into the category of the first image that has completed the clustering operation. For example: Determine the first image from which the human body characteristics are extracted, and complete the human body clustering operation between the second image and the first image based on the human body characteristics extracted from the first image and the human body characteristics extracted from the second image, and determine the difference between The image belongs to the first image in the same category. And according to the face clustering result, the category to which the first image belongs to the same category as the second image is determined, and the second image is further determined to belong to the category to which the first image belongs, and the third result is obtained. According to the face clustering result And the third result can complete the clustering operation for the image to be processed, and obtain the clustering result.
  • the human body clustering result may include a fourth result.
  • a human body clustering operation For any second image for which no facial features are extracted, according to the face clustering result and the second image
  • the human body features of and the human body features in the first image from which the human body features are extracted are subjected to a human body clustering operation to obtain a human body clustering result, including:
  • the fourth result is obtained.
  • At least one human body cluster center of an existing category can be obtained from an image database, and the similarity between the human body feature of the second image and the at least one human body cluster center of the existing category can be determined, and then the second image belongs to Existing category, cluster the second image into the existing category, and get the fourth result of the second image.
  • the clustering of the image to be processed can be completed Operate to get the clustering result.
  • images to be processed including 7 first images extracted from face images (including image to be processed 1, image to be processed 2, image to be processed 3, image to be processed 4, image to be processed 5 Human body features are extracted, images to be processed 6 and images to be processed 7 are not extracted from human features), 5 second images (images to be processed 8, images to be processed 9, images to be processed 10 are not extracted), and face images are not extracted.
  • the face clustering operation for the 7 first images can refer to the foregoing example, and this example will not be repeated here.
  • the first image image to be processed 1, image to be processed 2, image to be processed 3, image to be processed 4, image to be processed,
  • the human body features in the image 5) are clustered.
  • the third result of the human body clustering operation is that the image to be processed 9 and the image to be processed 2 belong to the same category (person 7), and the image to be processed 10 and the image to be processed 3 belong to the same category (person 7). In the same category (person 4), the image 12 to be processed and the image 4 to be processed belong to the same category (person 8).
  • the cluster centers of the 6 existing categories of human bodies can be obtained, respectively, based on the human body features extracted from the 2 second images and the 6 existing images.
  • the human body clustering center with categories performs the human body clustering operation, and the fourth result obtained is that the image 8 to be processed belongs to the person 1 and the image 11 to be processed belongs to the person 3.
  • the clustering operation for the 12 images to be processed is completed, and the clustering results for the 12 images to be processed are obtained: the image to be processed 1 and the image to be processed 8 belong to the same category (person 1), and the image to be processed 2 ,
  • the image to be processed 6 and the image to be processed 9 belong to the same category (person 7), the image to be processed 3, the image to be processed 5, and the image to be processed 10 belong to the same category (person 4),
  • the image to be processed 4 and the image to be processed 12 belongs to the same category (person 8), the image 7 to be processed belongs to one category (person 9), and the image 11 to be processed belongs to the person 3.
  • the foregoing method may further include:
  • the face cluster center and the human body cluster center of at least one category in the image database are updated according to the image to be processed.
  • the category to which at least one image to be processed belongs can be determined according to the clustering result, and then the at least one image to be processed is stored in the image database according to the corresponding category, and The face cluster center and the human body cluster center of at least one category are updated according to the image to be processed stored in at least one category.
  • Performing face clustering operations on the first image according to the extracted facial features includes: clustering the first image into existing categories according to the face clustering centers of the existing categories in the image database to obtain the first As a result, the face clustering operation is performed again for the first image that has not been clustered into the existing category, a new category is generated, and the second result is obtained, thereby determining the category to which at least one first image belongs, and according to the first image The result and the second result finally get the face clustering result.
  • For the second image perform a human body clustering operation based on the extracted human body characteristics, including: performing a human clustering operation on the second image and the first image from which the human body characteristics are extracted, and determining that the second image belongs to the same category as the first image.
  • One image the second image is clustered into the face category, and the third result is obtained; for the second image that is not clustered into the face category, the second image is clustered according to the human body cluster center of the existing category in the image data.
  • the images are clustered into existing categories to obtain a fourth result, so as to determine the category to which at least one second image belongs, and finally a human body clustering result is obtained according to the third result and the fourth result.
  • the human body clustering results can be merged with the face clustering results to obtain the final clustering results, the images to be processed are stored in the image database according to the clustering results, and the face clustering centers of each category are updated according to the images to be processed And the human cluster center.
  • the image to be processed can be clustered, and the image to be processed can be stored according to the category according to the clustering result.
  • the public security organ can be based on the relationship between the face and the human body. View clues and construct human-dimensional files more accurately, so as to better grasp the suspect's information, track the suspect's trajectory, conduct early warning and solve cases, etc.
  • the present disclosure also provides image processing devices, electronic equipment, computer-readable storage media, and programs, all of which can be used to implement any image processing method provided in the present disclosure.
  • image processing devices electronic equipment, computer-readable storage media, and programs, all of which can be used to implement any image processing method provided in the present disclosure.
  • the writing order of the steps does not mean a strict execution order but constitutes any limitation on the implementation process.
  • the specific execution order of each step should be based on its function and possibility.
  • the inner logic is determined.
  • Fig. 3 shows a schematic structural diagram of an image processing apparatus provided by an embodiment of the present disclosure. As shown in Fig. 3, the apparatus may include:
  • the extraction module 301 may be used to perform facial feature extraction and human body feature extraction on the image to be processed to obtain image features of the image to be processed.
  • the image features include facial features and/or human body features, and the image to be processed includes The first image and the second image;
  • the first clustering module 302 may be configured to perform a face clustering operation according to the extracted face features for the first image from which the face features are extracted, to obtain a face clustering result;
  • the second clustering module 303 may be used for the second image for which no facial features are extracted, according to the face clustering result, the human body features extracted from the second image, and the first human body feature extracted Perform a human body clustering operation on the human body features in an image to obtain a human body clustering result;
  • the third clustering module 304 may be used to obtain a clustering result for the image to be processed according to the face clustering result and the human body clustering result.
  • the functions or modules contained in the device provided in the embodiments of the present disclosure can be used to execute the methods described in the above method embodiments.
  • the functions or modules contained in the device provided in the embodiments of the present disclosure can be used to execute the methods described in the above method embodiments.
  • the face clustering result includes a first result
  • the first clustering module is further configured to:
  • the face clustering result further includes a second result
  • the first clustering module is further configured to:
  • the human body clustering result includes a third result
  • the second clustering module is further used for:
  • any of the second images perform a human body clustering operation based on the human body characteristics in the first image from which the human body characteristics are extracted and the human body characteristics in the second image to obtain a human body clustering sub-result ;
  • the second image is added to the category to which the first image belongs to the same human body category as the second image, to obtain the third result.
  • the human body clustering result further includes a fourth result
  • the second clustering module is further configured to:
  • the device further includes:
  • An adding module configured to add the to-be-processed image to an image database according to the clustering result
  • the update module is used to update at least one type of face cluster center and human body cluster center in the image database according to the to-be-processed image.
  • the functions or modules contained in the device provided in the embodiments of the present disclosure can be used to execute the methods described in the above method embodiments. For specific implementation, refer to the description of the above method embodiments. For brevity, here No longer
  • the embodiments of the present disclosure also provide a computer-readable storage medium on which computer program instructions are stored, and the computer program instructions implement the above-mentioned method when executed by a processor.
  • the computer-readable storage medium may be a non-volatile computer-readable storage medium.
  • An embodiment of the present disclosure also provides an electronic device, including: a processor; a memory for storing executable instructions of the processor; wherein the processor is configured as the above method.
  • the embodiment of the present disclosure also proposes a computer program, including computer readable code, and when the computer readable code runs in an electronic device, a processor in the electronic device executes for realizing the above-mentioned image processing method.
  • the electronic device can be provided as a terminal, server or other form of device.
  • Fig. 4 is a block diagram showing an electronic device 800 according to an exemplary embodiment.
  • the electronic device 800 may be a mobile phone, a computer, a digital broadcasting terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, and other terminals.
  • the electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, and a sensor component 814 , And communication component 816.
  • the processing component 802 generally controls the overall operations of the electronic device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations.
  • the processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the foregoing method.
  • the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components.
  • the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.
  • the memory 804 is configured to store various types of data to support operations in the electronic device 800. Examples of these data include instructions for any application or method operating on the electronic device 800, contact data, phone book data, messages, pictures, videos, etc.
  • the memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable and Programmable read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.
  • SRAM static random access memory
  • EEPROM electrically erasable programmable read-only memory
  • EPROM erasable and Programmable read only memory
  • PROM programmable read only memory
  • ROM read only memory
  • magnetic memory flash memory
  • flash memory magnetic disk or optical disk.
  • the power supply component 806 provides power for various components of the electronic device 800.
  • the power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 800.
  • the multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user.
  • the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user.
  • the touch panel includes one or more touch sensors to sense touch, sliding, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure related to the touch or slide operation.
  • the multimedia component 808 includes a front camera and/or a rear camera. When the electronic device 800 is in an operation mode, such as a shooting mode or a video mode, the front camera and/or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have focal length and optical zoom capabilities.
  • the audio component 810 is configured to output and/or input audio signals.
  • the audio component 810 includes a microphone (MIC), and when the electronic device 800 is in an operation mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal.
  • the received audio signal may be further stored in the memory 804 or transmitted via the communication component 816.
  • the audio component 810 further includes a speaker for outputting audio signals.
  • the I/O interface 812 provides an interface between the processing component 802 and a peripheral interface module.
  • the above-mentioned peripheral interface module may be a keyboard, a click wheel, a button, and the like. These buttons may include, but are not limited to: home button, volume button, start button, and lock button.
  • the sensor component 814 includes one or more sensors for providing the electronic device 800 with various aspects of state evaluation.
  • the sensor component 814 can detect the on/off status of the electronic device 800 and the relative positioning of the components.
  • the component is the display and the keypad of the electronic device 800.
  • the sensor component 814 can also detect the electronic device 800 or the electronic device 800.
  • the position of the component changes, the presence or absence of contact between the user and the electronic device 800, the orientation or acceleration/deceleration of the electronic device 800, and the temperature change of the electronic device 800.
  • the sensor component 814 may include a proximity sensor configured to detect the presence of nearby objects when there is no physical contact.
  • the sensor component 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications.
  • the sensor component 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
  • the communication component 816 is configured to facilitate wired or wireless communication between the electronic device 800 and other devices.
  • the electronic device 800 can access a wireless network based on a communication standard, such as WiFi, 2G, or 3G, or a combination thereof.
  • the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel.
  • the communication component 816 further includes a near field communication (NFC) module to facilitate short-range communication.
  • the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.
  • RFID radio frequency identification
  • IrDA infrared data association
  • UWB ultra-wideband
  • Bluetooth Bluetooth
  • the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field-available A programmable gate array (FPGA), controller, microcontroller, microprocessor, or other electronic components are implemented to implement the above methods.
  • ASIC application-specific integrated circuits
  • DSP digital signal processors
  • DSPD digital signal processing devices
  • PLD programmable logic devices
  • FPGA field-available A programmable gate array
  • controller microcontroller, microprocessor, or other electronic components are implemented to implement the above methods.
  • a non-volatile computer-readable storage medium such as the memory 804 including computer program instructions, which can be executed by the processor 820 of the electronic device 800 to complete the foregoing method.
  • Fig. 5 is a block diagram showing an electronic device 1900 according to an exemplary embodiment.
  • the electronic device 1900 may be provided as a server. 5
  • the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932, for storing instructions executable by the processing component 1922, such as application programs.
  • the application program stored in the memory 1932 may include one or more modules each corresponding to a set of instructions.
  • the processing component 1922 is configured to execute instructions to perform the above-described methods.
  • the electronic device 1900 may also include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input output (I/O) interface 1958 .
  • the electronic device 1900 can operate based on an operating system stored in the memory 1932, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM or the like.
  • a non-volatile computer-readable storage medium is also provided, such as the memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the electronic device 1900 to complete the foregoing method.
  • the present disclosure may be a system, method and/or computer program product.
  • the computer program product may include a computer-readable storage medium loaded with computer-readable program instructions for enabling a processor to implement various aspects of the present disclosure.
  • the computer-readable storage medium may be a tangible device that can hold and store instructions used by the instruction execution device.
  • the computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
  • Non-exhaustive list of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) Or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital versatile disk (DVD), memory stick, floppy disk, mechanical encoding device, such as a printer with instructions stored thereon
  • RAM random access memory
  • ROM read-only memory
  • EPROM erasable programmable read-only memory
  • flash memory flash memory
  • SRAM static random access memory
  • CD-ROM compact disk read-only memory
  • DVD digital versatile disk
  • memory stick floppy disk
  • mechanical encoding device such as a printer with instructions stored thereon
  • the computer-readable storage medium used here is not interpreted as the instantaneous signal itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (for example, light pulses through fiber optic cables), or through wires Transmission of electrical signals.
  • the computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to various computing/processing devices, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and/or a wireless network.
  • the network may include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers.
  • the network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network, and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing/processing device .
  • the computer program instructions used to perform the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or in one or more programming languages.
  • Source code or object code written in any combination, the programming language includes object-oriented programming languages such as Smalltalk, C++, etc., and conventional procedural programming languages such as "C" language or similar programming languages.
  • Computer-readable program instructions can be executed entirely on the user's computer, partly on the user's computer, executed as a stand-alone software package, partly on the user's computer and partly executed on a remote computer, or entirely on the remote computer or server carried out.
  • the remote computer can be connected to the user's computer through any kind of network-including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect to the user's computer) connection).
  • LAN local area network
  • WAN wide area network
  • an electronic circuit such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), can be customized by using the status information of the computer-readable program instructions.
  • FPGA field programmable gate array
  • PDA programmable logic array
  • the computer-readable program instructions are executed to realize various aspects of the present disclosure.
  • These computer-readable program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine that makes these instructions when executed by the processor of the computer or other programmable data processing device , A device that implements the functions/actions specified in one or more blocks in the flowcharts and/or block diagrams is produced. It is also possible to store these computer-readable program instructions in a computer-readable storage medium. These instructions make computers, programmable data processing apparatuses, and/or other devices work in a specific manner. Thus, the computer-readable medium storing the instructions includes An article of manufacture, which includes instructions for implementing various aspects of the functions/actions specified in one or more blocks in the flowcharts and/or block diagrams.
  • each block in the flowchart or block diagram may represent a module, program segment, or part of an instruction, and the module, program segment, or part of an instruction contains one or more components for realizing the specified logical function.
  • Executable instructions may also occur in a different order than the order marked in the drawings. For example, two consecutive blocks can actually be executed substantially in parallel, or they can sometimes be executed in the reverse order, depending on the functions involved.
  • each block in the block diagram and/or flowchart, and the combination of the blocks in the block diagram and/or flowchart can be implemented by a dedicated hardware-based system that performs the specified functions or actions Or it can be realized by a combination of dedicated hardware and computer instructions.

Abstract

一种图像处理方法及装置、电子设备和存储介质，所述方法包括：对待处理图像进行人脸特征提取及人体特征提取，得到待处理图像的图像特征，图像特征包括人脸特征和/或人体特征，所述待处理图像包括第一图像及第二图像（S11）；针对提取到人脸特征的所述第一图像，根据提取到的人脸特征进行人脸聚类操作，得到人脸聚类结果（S12）；针对未提取到人脸特征的第二图像，根据人脸聚类结果、第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果（S13）；根据人脸聚类结果及人体聚类结果得到针对待处理图像的聚类结果（S14）。该方法可在保证聚类结果的准确性的同时，提高了聚类结果的召回率。

Description

图像处理方法及装置、电子设备和存储介质
本申请要求在2019年8月30日提交中国专利局、申请号为201910818028.9、发明名称为“图像处理方法及装置、电子设备和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本公开涉及人工智能技术领域，尤其涉及一种图像处理方法及装置、电子设备和存储介质。
背景技术
随着相关技术的发展，人脸检索得到了广泛应用，特别是在公安行业破案时，需要根据未确认身份的嫌疑人图像在海量人像库中进行检索，基于此需要建立一人一档的信息库，该信息库中同一人的图像归属于同一类别。
相关技术中可以采用人脸聚类的方式对图像进行聚类，以建立上述信息库。
发明内容
本公开提出了一种图像处理技术方案。
根据本公开的一方面，提供了一种图像处理方法，包括：
对待处理图像进行人脸特征提取及人体特征提取，得到所述待处理图像的图像特征，所述图像特征包括人脸特征和/或人体特征，所述待处理图像包括第一图像及第二图像；
针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果；
针对未提取到人脸特征的所述第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果；
根据所述人脸聚类结果及所述人体聚类结果得到针对所述待处理图像的聚类结果。
在一种可能的实现方式中，所述人脸聚类结果包括第一结果，所述针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果，包括：
获取图像数据库中的至少一个已有类别的人脸聚类中心；
根据所述至少一个已有类别的人脸聚类中心及所述第一图像中提取到的所述人脸特征进行人脸聚类，将所述第一图像聚类至所述已有类别中，得到所述第一图像的所述第一结果。
在一种可能的实现方式中，所述人脸聚类结果还包括第二结果，所述针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果，还包括：
对未聚类到所述已有类别中的所述第一图像进行人脸聚类操作，得到所述第一图像的所述第二结果。
在一种可能的实现方式中，所述人体聚类结果包括第三结果，所述针对未提取到人脸特征的第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果，包括：
针对任一所述第二图像，根据提取出人体特征的所述第一图像中的所述人体特征及所述第二图像中的所述人体特征进行人体聚类操作，得到人体聚类子结果；
根据所述人体聚类子结果确定与所述第二图像归属于同一人体类别的所述第一图像；
根据所述人脸聚类结果，将所述第二图像添加至与所述第二图像归属于同一人体类别的所述第一图像所属的类别中，得到所述第三结果。
在一种可能的实现方式中，所述人体聚类结果还包括第四结果，所述针对任一未提取到人脸特征的第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的所述第一图像中的人体特征进行人体聚类操作，得到人体聚类结果，包括：
获取图像数据库中至少一个已有类别的人体聚类中心；
对未聚类到人脸类别中的所述第二图像，根据所述第二图像中的所述人体特征及所述至少一个已有类别的所述人体聚类中心进行人体聚类操作，将所述第二图像聚类至所述已有类别中，得到所述第四结果。
在一种可能的实现方式中，所述方法还包括：
根据所述聚类结果将所述待处理图像添加至图像数据库中；
根据所述待处理图像更新所述图像数据库中至少一个类别的人脸聚类中心及人体聚类中心。
根据本公开的一方面，提供了一种图像处理装置，包括：
提取模块，用于对待处理图像进行人脸特征提取及人体特征提取，得到所述待处理图像的图像特征，所述图像特征包括人脸特征和/或人体特征，所述待处理图像包括第一图像及第二图像；
第一聚类模块，用于针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果；
第二聚类模块，用于针对未提取到人脸特征的所述第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果；
第三聚类模块，用于根据所述人脸聚类结果及所述人体聚类结果得到针对所述待处理图像的聚类结果。
在一种可能的实现方式中，所述人脸聚类结果包括第一结果，所述第一聚类模块，还用于：
获取图像数据库中的至少一个已有类别的人脸聚类中心；
根据所述至少一个已有类别的人脸聚类中心及所述第一图像中提取到的所述人脸特征进行人脸聚类，将所述第一图像聚类至所述已有类别中，得到所述第一图像的所述第一结果。
在一种可能的实现方式中，所述人脸聚类结果还包括第二结果，所述第一聚类模块，还用于：
对未聚类到所述已有类别中的所述第一图像进行人脸聚类操作，得到所述第一图像的所述第二结果。
在一种可能的实现方式中，所述人体聚类结果包括第三结果，所述第二聚类模块，还用于：
针对任一所述第二图像，根据提取出人体特征的所述第一图像中的所述人体特征及所述第二图像中的所述人体特征进行人体聚类操作，得到人体聚类子结果；
根据所述人体聚类子结果确定与所述第二图像归属于同一人体类别的所述第一图像；
根据所述人脸聚类结果，将所述第二图像添加至与所述第二图像归属于同一人体类别的所述第一图像所属的类别中，得到所述第三结果。
在一种可能的实现方式中，所述人体聚类结果还包括第四结果，所述第二聚类模块，还用于：
获取图像数据库中至少一个已有类别的人体聚类中心；
对未聚类到人脸类别中的所述第二图像，根据所述第二图像中的所述人体特征及所述至少一个已有类别的所述人体聚类中心进行人体聚类操作，将所述第二图像聚类至所述已有类别中，得到所述第四结果。
在一种可能的实现方式中，所述装置还包括：
添加模块，用于根据所述聚类结果将所述待处理图像添加至图像数据库中；
更新模块，用于根据所述待处理图像更新所述图像数据库中至少一个类别的人脸聚类中心及人体聚类中心。
根据本公开的一方面，提供了一种电子设备，包括：
处理器；
用于存储处理器可执行指令的存储器；
其中，所述处理器被配置为：执行上述图像处理方法。
根据本公开的一方面，提供了一种计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现上述图像处理方法。
根据本公开的一方面，提供了一种计算机程序，包括计算机可读代码，当所述计算机可读代码在电子设备中运行时，所述电子设备中的处理器执行用于实现上述图像处理方法。
这样，根据本公开实施例提供的图像处理方法及装置，可以通过人脸聚类的方式对能够提取出人脸特征的第一图像进行人脸聚类，确定第一图像所属的类别，得到人脸聚类结果。通过人体聚类的方式将未提取出人脸特征的第二图像与第一图像进行人体聚类，进而根据第一图像的人脸聚类结果确定第二图像所属的类别，以此得到针对待处理图像的聚类结果，由于本公开实施例采用人脸聚类结合人体聚类的方式，因此在保证聚类结果的准确性的同时，提高了聚类结果的召回率。
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，而非限制本公开。
根据下面参考附图对示例性实施例的详细说明，本公开的其它特征及方面将变得清楚。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，这些附图示出了符合本公开的实施例，并与说明书一起用于说明本公开的技术方案。
图1示出根据本公开实施例的图像处理方法的流程图；
图2示出根据本公开实施例的图像处理方法的示意图；
图3示出本公开实施例提供的一种图像处理装置的结构示意图；
图4是根据一示例性实施例示出的一种电子设备800的框图；
图5是根据一示例性实施例示出的一种电子设备1900的框图。
具体实施方式
以下将参考附图详细说明本公开的各种示例性实施例、特征和方面。附图中相同的附图标记表示功能相同或相似的元件。尽管在附图中示出了实施例的各种方面，但是除非特别指出，不必按比例绘制附图。
在这里专用的词“示例性”意为“用作例子、实施例或说明性”。这里作为“示例性”所说明的任何实施例不必解释为优于或好于其它实施例。
本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中术语“至少一种”表示多种中的任意一种或多种中的至少两种的任意组合，例如，包括A、B、C中的至少一种，可以表示包括从A、B和C构成的集合中选择的任意一个或多个元素。
另外，为了更好地说明本公开，在下文的具体实施方式中给出了众多的具体细节。本领域技术人员应当理解，没有某些具体细节，本公开同样可以实施。在一些实例中，对于本领域技术人员熟知的方法、手段、元件和电路未作详细描述，以便于凸显本公开的主旨。
图1示出根据本公开实施例的图像处理方法的流程图。该图像处理方法可以由终端设备或其它处理设备执行，其中，终端设备可以为用户设备(User Equipment，UE)、移动设备、用户终端、终端、蜂窝电话、无绳电话、个人数字处理(Personal Digital Assistant，PDA)、手持设备、计算设备、车载设备、可穿戴设备等。其它处理设备可为服务器或云端服务器等。在一些可能的实现方式中，该图像处理方法可以通过处理器调用存储器中存储的计算机可读指令的方式来实现。
如图1所示，所述方法包括：
步骤S11、对待处理图像进行人脸特征提取及人体特征提取，得到所述待处理图像的图像特征，所述图像特征包括人脸特征和/或人体特征，所述待处理图像包括第一图像和第二图像。
举例来说，所述待处理图像可以是由图像采集设备(例如摄像头)采集的图像，也可以是直接输入的已保存的图像或者视频帧。
例如，可使用卷积神经网络等神经网络对待处理图像进行人脸特征提取及人体特征提取，得到待处理图像的图像特征(人脸特征和/或人体特征)，例如：对多张待处理图像进行人脸特征提取及人体特征提取后，部分待处理图像采集到了人脸特征，可以确定该部分待处理图像为第一图像，该第一图像还可以分为仅采集到人脸特征未采集到人体特征的图像，及，既采集到人脸特征又采集到人体特征的图像，另一部分未采集到人脸特征但采集到了人体特征的待处理图像可以确定为第二图像。本公开对神经网络的类型不作限制，对提取人脸特征和人体特征的方式不做限制。
在一种可能的实现方式中，人脸特征可以是根据人脸关键点确定的特征信息，例如，五官的位置、形状等，还可包括肤色等信息。人体特征可以是根据人体关键点确定的特征信息，例如，身高、体型、腿长、臂长等，还可包括衣物的款式、颜色等信息。
步骤S12、针对提取到所述人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果。
举例来说，可以确定提取到人脸特征的待处理图像为第一图像，确定未提取到人脸特征但提取到人体特征的待处理图像为第二图像。可以根据该第一图像中提取到的人脸特征进行人脸聚类操作，例 如：将多张第一图像进行人脸聚类，得到人脸聚类结果，或者在当前聚类操作之前，预先已经根据历史图像进行过聚类，图像数据库中已按照已有类别存储图像，可以将第一图像聚类到已有类别中去，并对无法聚类到已有类别中去的第一图像重新进行聚类，得到人脸聚类结果。
示例性的，上述人脸聚类操作可以采用K-MEANS算法、K-MEDOIDS算法、CLARANS算法等任一聚类方式，本公开对于人脸聚类操作所采用的聚类方式不作具体限定。
步骤S13、针对未提取到人脸特征的所述第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果。
步骤S14、根据所述人脸聚类结果及所述人体聚类结果得到针对所述待处理图像的聚类结果。
针对既提取出人脸特征又提取出人体特征的第一图像及未提取出人脸图像的第二图像，可以根据提取出的人体特征执行两者间的人体聚类操作，确定与第二图像归为一类的第一图像，得到人体聚类结果。
进一步的，可以将人脸聚类结果及人体聚类结果进行融合，得到针对待处理图像的聚类结果。
示例性的，上述人体聚类操作可以采用K-MEANS算法、K-MEDOIDS算法、CLARANS算法等任一聚类方式，本公开对于人体聚类操作所采用的聚类方式不作具体限定。
这样，根据本公开实施例提供的图像处理方法，可以通过人脸聚类的方式对能够提取出人脸特征的第一图像进行人脸聚类，确定第一图像所属的类别，得到人脸聚类结果。通过人体聚类的方式将未提取出人脸特征的第二图像与第一图像进行人体聚类，进而根据第一图像的人脸聚类结果确定第二图像所属的类别，以此得到针对待处理图像的聚类结果，由于本公开实施例采用人脸聚类结合人体聚类的方式，因此在保证聚类结果的准确性的同时，提高了聚类结果的召回率。
在一种可能的实现方式中，上述人脸聚类结果可以包括第一结果，上述针对提取到人脸特征的所述第一图像，根据提取到的人脸特征进行人脸聚类操作，得到人脸聚类结果，可以包括：
获取图像数据库中的至少一个已有类别的人脸聚类中心；
根据所述至少一个已有类别的所述人脸聚类中心及所述第一图像中提取到的所述人脸特征进行人脸聚类，将所述第一图像聚类至所述已有类别中，得到所述第一图像的第一结果。
举例来说，在当前聚类操作之前预先已经根据历史图像进行过聚类，并且已将历史图像按照聚类结果所确定的类别存储于图像数据库中，该图像数据库中还存储有聚类操作得到的至少一个已有类别的人脸聚类中心和人体聚类中心，其中，针对任一已有类别，人脸聚类中心可以为该已有类别对应的图像中提取到的人脸特征的均值，人体聚类中心可以为该已有类别对应的图像中提取到的人体特征的均值。
可以从图像数据库中获取至少一个已有类别的人脸聚类中心，确定第一图像的人脸特征与至少一个已有类别的人脸聚类中心的相似度，进而根据该相似度确定第一图像所属的类别，将第一图像聚类到已有类别中去，得到第一图像的第一结果。
在一种可能的实现方式中，所述人脸聚类结果可以包括第二结果，上述针对提取到人脸特征的第一图像，根据提取到的人脸特征进行人脸聚类操作，得到人脸聚类结果，还包括：
对未聚类到所述已有类别中的所述第一图像进行人脸聚类操作，得到第一图像的第二结果。
举例来说，针对未聚类至已有类别中的至少一张第一图像，根据第一图像中提取出的人脸特征再 次进行人脸聚类操作，将该多张第一图像聚类成新的类别，得到第一图像的第二结果。
示例性的，当前存在7张第一图像(待处理图像1，待处理图像2，待处理图像3，待处理图像4，待处理图像5，待处理图像6，待处理图像7)，图像数据库中存在6个已有类别(人物1，人物2，人物3，人物4，人物5，人物6)，可以分别获取6个已有类别的人脸聚类中心，分别根据7张第一图像中提取出的人脸特征及6个已有类别的人脸聚类中心执行人脸聚类操作，得到第一结果为待处理图像1属于人物1，待处理图像3和待处理图像5属于人物4。针对未聚类到已有类别的待处理图像2、待处理图像4、待处理图像6、待处理图像7，根据提取出的人脸特征进行人脸聚类操作，得到第二结果为：待处理图像2和待处理图像6归属于一类(人物7)、待处理图像4属于一类(人物8)、待处理图像7属于一类(人物9)，第一结果与第二结果合并得到了针对第一图像的人脸聚类结果。
在一种可能的实现方式中，人体聚类结果可以包括第三结果，在完成人脸聚类操作后，上述针对未提取到人脸特征的所述第二图像，根据上述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果，包括：
针对任一所述第二图像，根据提取出人体特征的所述第一图像中的人体特征及该第二图像中的人体特征进行人体聚类操作，得到人体聚类子结果；
根据所述人体聚类结果确定与所述第二图像归属于同一人体类别的所述第一图像；
根据所述人脸聚类结果，将所述第二图像添加至与所述第二图像归属于同一人体类别的所述第一图像所属的类别中，得到第三结果。
针对任一第二图像，可以将其聚类至已完成聚类操作的第一图像所属的类别中。例如：确定提取出人体特征的第一图像，根据该第一图像提取出的人体特征及第二图像提取的人体特征，完成第二图像与该第一图像的人体聚类操作，确定与第二图像归属于同一类别的第一图像。并根据人脸聚类结果，确定与第二图像归属同一类的第一图像所属的类别，进一步的确定第二图像属于该第一图像所属的类别，得到第三结果，根据人脸聚类结果及第三结果可以完成针对待处理图像的聚类操作，得到聚类结果。
在一种可能的实现方式中，人体聚类结果可以包括第四结果，上述针对任一未提取到人脸特征的第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的所述第一图像中的人体特征进行人体聚类操作，得到人体聚类结果，包括：
获取图像数据库中至少一个已有类别的人体聚类中心；
对未聚类到人脸类别中的所述第二图像，根据所述第二图像中的人体特征及所述至少一个已有类别的人体聚类中心进行人体聚类，将所述第二图像聚类至已有类别中，得到第四结果。
举例来说，可以从图像数据库中获取至少一个已有类别的人体聚类中心，确定第二图像的人体特征与至少一个已有类别的人体聚类中心的相似度，进而确定第二图像所属的已有类别，将第二图像聚类到该已有类别中去，得到第二图像的第四结果，根据人脸聚类结果、第三结果及第四结果可以完成针对待处理图像的聚类操作，得到聚类结果。
示例性的，当前存在12张待处理图像，其中包括7张提取到人脸图像的第一图像(其中待处理图像1、待处理图像2、待处理图像3、待处理图像4、待处理图像5提取到人体特征，待处理图像6及待处理图像7未提取到人体特征)，5张未提取到人脸图像的第二图像(待处理图像8，待处理图像9，待处 理图像10，待处理图像11，待处理图像12)。针对7张第一图像的人脸聚类操作可以参照前述示例，本示例在此不再赘述。
针对5张第二图像，分别根据第二图像提取到的人体特征及上述提取到人体特征的第一图像(待处理图像1、待处理图像2、待处理图像3、待处理图像4、待处理图像5)中的人体特征进行人体聚类操作，人体聚类操作得到的第三结果是待处理图像9与待处理图像2属于同一类别(人物7)，待处理图像10与待处理图像3属于同一类别(人物4)，待处理图像12与待处理图像4属于同一类别(人物8)。
对于未聚类到人脸类别的待处理图像8和待处理图像11，可以分别获取6个已有类别的人体聚类中心，分别根据2张第二图像中提取出的人体特征及6个已有类别的人体聚类中心执行人体聚类操作，得到的第四结果为待处理图像8属于人物1，待处理图像11属于人物3。
至此，完成了针对12张待处理图像的聚类操作，得到了针对12张待处理图像的聚类结果：待处理图像1与待处理图像8归属于同一类(人物1)，待处理图像2、待处理图像6及待处理图像9归属于同一类(人物7)，待处理图像3、待处理图像5及待处理图像10归属于同一类(人物4)，待处理图像4与待处理图像12归属于同一类(人物8)，待处理图像7属于一类(人物9)、待处理图像11属于人物3。
在一种可能的实现方式中，上述方法还可以包括：
根据所述聚类结果将所述待处理图像添加至图像数据库中；
根据所述待处理图像更新所述图像数据库中至少一个类别的人脸聚类中心及人体聚类中心。
举例来说，在完成对待处理图像的聚类操作后，可以根据聚类结果确定至少一个待处理图像所属的类别，进而将至少一个待处理图像在图像数据库中按照对应的类别进行存储，并可以根据存储进至少一个类别的待处理图像进行至少一个类别的人脸聚类中心及人体聚类中心的更新。
为使本领域技术人员更好的理解本申请实施例，以下可以结合图2所示的示例对本申请实施例加以说明。
针对待处理图像进行人脸特征提取及人体特征提取，确定提取到人脸特征的待处理图像为第一图像，未提取到人脸特征的图像为第二图像。
针对第一图像根据其提取到的人脸特征进行人脸聚类操作，包括：根据图像数据库中已有类别的人脸聚类中心将第一图像聚类至已有类别中去，得到第一结果，对于未聚类至已有类别中去的第一图像再次进行人脸聚类操作，生成新的类别，得到第二结果，以此确定了至少一个第一图像所属的类别，根据第一结果及第二结果最终得到人脸聚类结果。
针对第二图像，根据其提取到的人体特征进行人体聚类操作，包括：将第二图像与提取到人体特征的第一图像执行人类聚类操作，确定与第二图像归属于同一类的第一图像，将第二图像聚类至人脸类别中去，得到第三结果；针对未聚类到人脸类别中的第二图像，根据图像数据中已有类别的人体聚类中心将第二图像聚类至已有类别中去，得到第四结果，以此确定至少一个第二图像所属的类别，根据第三结果及第四结果最终得到人体聚类结果。
可以将人体聚类结果与人脸聚类结果进行融合，得到最终的聚类结果，将待处理图像按照聚类结果存储至图像数据库中，并根据待处理图像更新各类别的人脸聚类中心及人体聚类中心。
示例性的，根据本公开实施例提供的图像处理方法，可以将待处理图像进行聚类操作，并按照聚类结果将待处理图像按照所属类别进行存储，例如公安机关可以基于人脸与人体的视图线索，更为精 准的构建以人为维度的档案，从而更好的掌握嫌疑人信息、追踪嫌疑人轨迹，进行预警与破案等。
可以理解，本公开提及的上述各个方法实施例，在不违背原理逻辑的情况下，均可以彼此相互结合形成结合后的实施例，限于篇幅，本公开不再赘述。
此外，本公开还提供了图像处理装置、电子设备、计算机可读存储介质、程序，上述均可用来实现本公开提供的任一种图像处理方法，相应技术方案和描述和参见方法部分的相应记载，不再赘述。
本领域技术人员可以理解，在具体实施方式的上述方法中，各步骤的撰写顺序并不意味着严格的执行顺序而对实施过程构成任何限定，各步骤的具体执行顺序应当以其功能和可能的内在逻辑确定。
图3示出本公开实施例提供的一种图像处理装置的结构示意图，如图3所示，该装置可以包括：
提取模块301，可以用于对待处理图像进行人脸特征提取及人体特征提取，得到所述待处理图像的图像特征，所述图像特征包括人脸特征和/或人体特征，所述待处理图像包括第一图像及第二图像；
第一聚类模块302，可以用于针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果；
第二聚类模块303，可以用于针对未提取到人脸特征的所述第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果；
第三聚类模块304，可以用于根据所述人脸聚类结果及所述人体聚类结果得到针对所述待处理图像的聚类结果。
在本公开的一些实施例中，本公开实施例提供的装置具有的功能或包含的模块可以用于执行上文方法实施例描述的方法，其具体实现和技术效果可参照上文方法实施例的描述，为了简洁，这里不再赘述。
在一种可能的实现方式中，所述人脸聚类结果包括第一结果，所述第一聚类模块，还用于：
获取图像数据库中的至少一个已有类别的人脸聚类中心；
根据所述至少一个已有类别的人脸聚类中心及所述第一图像中提取到的所述人脸特征进行人脸聚类，将所述第一图像聚类至所述已有类别中，得到所述第一图像的所述第一结果。
在一种可能的实现方式中，所述人脸聚类结果还包括第二结果，所述第一聚类模块，还用于：
对未聚类到所述已有类别中的所述第一图像进行人脸聚类操作，得到所述第一图像的所述第二结果。
在一种可能的实现方式中，所述人体聚类结果包括第三结果，所述第二聚类模块，还用于：
针对任一所述第二图像，根据提取出人体特征的所述第一图像中的所述人体特征及所述第二图像中的所述人体特征进行人体聚类操作，得到人体聚类子结果；
根据所述人体聚类子结果确定与所述第二图像归属于同一人体类别的所述第一图像；
根据所述人脸聚类结果，将所述第二图像添加至与所述第二图像归属于同一人体类别的所述第一图像所属的类别中，得到所述第三结果。
在一种可能的实现方式中，所述人体聚类结果还包括第四结果，所述第二聚类模块，还用于：
获取图像数据库中至少一个已有类别的人体聚类中心；
对未聚类到人脸类别中的所述第二图像，根据所述第二图像中的所述人体特征及所述至少一个已有类别的所述人体聚类中心进行人体聚类操作，将所述第二图像聚类至所述已有类别中，得到所述第四结果。
在一种可能的实现方式中，所述装置还包括：
添加模块，用于根据所述聚类结果将所述待处理图像添加至图像数据库中；
更新模块，用于根据所述待处理图像更新所述图像数据库中至少一个类别的人脸聚类中心及人体聚类中心。在一些实施例中，本公开实施例提供的装置具有的功能或包含的模块可以用于执行上文方法实施例描述的方法，其具体实现可以参照上文方法实施例的描述，为了简洁，这里不再赘述
本公开实施例还提出一种计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现上述方法。计算机可读存储介质可以是非易失性计算机可读存储介质。
本公开实施例还提出一种电子设备，包括：处理器；用于存储处理器可执行指令的存储器；其中，所述处理器被配置为上述方法。
本公开实施例还提出一种计算机程序，包括计算机可读代码，当所述计算机可读代码在电子设备中运行时，所述电子设备中的处理器执行用于实现上述图像处理方法。
电子设备可以被提供为终端、服务器或其它形态的设备。
图4是根据一示例性实施例示出的一种电子设备800的框图。例如，电子设备800可以是移动电话，计算机，数字广播终端，消息收发设备，游戏控制台，平板设备，医疗设备，健身设备，个人数字助理等终端。
参照图4，电子设备800可以包括以下一个或多个组件：处理组件802，存储器804，电源组件806，多媒体组件808，音频组件810，输入/输出(I/O)的接口812，传感器组件814，以及通信组件816。
处理组件802通常控制电子设备800的整体操作，诸如与显示，电话呼叫，数据通信，相机操作和记录操作相关联的操作。处理组件802可以包括一个或多个处理器820来执行指令，以完成上述的方法的全部或部分步骤。此外，处理组件802可以包括一个或多个模块，便于处理组件802和其他组件之间的交互。例如，处理组件802可以包括多媒体模块，以方便多媒体组件808和处理组件802之间的交互。
存储器804被配置为存储各种类型的数据以支持在电子设备800的操作。这些数据的示例包括用于在电子设备800上操作的任何应用程序或方法的指令，联系人数据，电话簿数据，消息，图片，视频等。存储器804可以由任何类型的易失性或非易失性存储设备或者它们的组合实现，如静态随机存取存储器(SRAM)，电可擦除可编程只读存储器(EEPROM)，可擦除可编程只读存储器(EPROM)，可编程只读存储器(PROM)，只读存储器(ROM)，磁存储器，快闪存储器，磁盘或光盘。
电源组件806为电子设备800的各种组件提供电力。电源组件806可以包括电源管理系统，一个或多个电源，及其他与为电子设备800生成、管理和分配电力相关联的组件。
多媒体组件808包括在所述电子设备800和用户之间的提供一个输出接口的屏幕。在一些实施例中，屏幕可以包括液晶显示器(LCD)和触摸面板(TP)。如果屏幕包括触摸面板，屏幕可以被实现为触摸屏，以接收来自用户的输入信号。触摸面板包括一个或多个触摸传感器以感测触摸、滑动和触摸面板上的手势。所述触摸传感器可以不仅感测触摸或滑动动作的边界，而且还检测与所述触摸或滑动操作相关的持续时间和压力。在一些实施例中，多媒体组件808包括一个前置摄像头和/或后置摄像 头。当电子设备800处于操作模式，如拍摄模式或视频模式时，前置摄像头和/或后置摄像头可以接收外部的多媒体数据。每个前置摄像头和后置摄像头可以是一个固定的光学透镜系统或具有焦距和光学变焦能力。
音频组件810被配置为输出和/或输入音频信号。例如，音频组件810包括一个麦克风(MIC)，当电子设备800处于操作模式，如呼叫模式、记录模式和语音识别模式时，麦克风被配置为接收外部音频信号。所接收的音频信号可以被进一步存储在存储器804或经由通信组件816发送。在一些实施例中，音频组件810还包括一个扬声器，用于输出音频信号。
I/O接口812为处理组件802和外围接口模块之间提供接口，上述外围接口模块可以是键盘，点击轮，按钮等。这些按钮可包括但不限于：主页按钮、音量按钮、启动按钮和锁定按钮。
传感器组件814包括一个或多个传感器，用于为电子设备800提供各个方面的状态评估。例如，传感器组件814可以检测到电子设备800的打开/关闭状态，组件的相对定位，例如所述组件为电子设备800的显示器和小键盘，传感器组件814还可以检测电子设备800或电子设备800一个组件的位置改变，用户与电子设备800接触的存在或不存在，电子设备800方位或加速/减速和电子设备800的温度变化。传感器组件814可以包括接近传感器，被配置用来在没有任何的物理接触时检测附近物体的存在。传感器组件814还可以包括光传感器，如CMOS或CCD图像传感器，用于在成像应用中使用。在一些实施例中，该传感器组件814还可以包括加速度传感器，陀螺仪传感器，磁传感器，压力传感器或温度传感器。
通信组件816被配置为便于电子设备800和其他设备之间有线或无线方式的通信。电子设备800可以接入基于通信标准的无线网络，如WiFi，2G或3G，或它们的组合。在一个示例性实施例中，通信组件816经由广播信道接收来自外部广播管理系统的广播信号或广播相关信息。在一个示例性实施例中，所述通信组件816还包括近场通信(NFC)模块，以促进短程通信。例如，在NFC模块可基于射频识别(RFID)技术，红外数据协会(IrDA)技术，超宽带(UWB)技术，蓝牙(BT)技术和其他技术来实现。
在示例性实施例中，电子设备800可以被一个或多个应用专用集成电路(ASIC)、数字信号处理器(DSP)、数字信号处理设备(DSPD)、可编程逻辑器件(PLD)、现场可编程门阵列(FPGA)、控制器、微控制器、微处理器或其他电子元件实现，用于执行上述方法。
在示例性实施例中，还提供了一种非易失性计算机可读存储介质，例如包括计算机程序指令的存储器804，上述计算机程序指令可由电子设备800的处理器820执行以完成上述方法。
图5是根据一示例性实施例示出的一种电子设备1900的框图。例如，电子设备1900可以被提供为一服务器。参照图5，电子设备1900包括处理组件1922，其进一步包括一个或多个处理器，以及由存储器1932所代表的存储器资源，用于存储可由处理组件1922的执行的指令，例如应用程序。存储器1932中存储的应用程序可以包括一个或一个以上的每一个对应于一组指令的模块。此外，处理组件1922被配置为执行指令，以执行上述方法。
电子设备1900还可以包括一个电源组件1926被配置为执行电子设备1900的电源管理，一个有线或无线网络接口1950被配置为将电子设备1900连接到网络，和一个输入输出(I/O)接口1958。电子设备1900可以操作基于存储在存储器1932的操作系统，例如Windows ServerTM，Mac OS XTM，UnixTM, LinuxTM，FreeBSDTM或类似。
在示例性实施例中，还提供了一种非易失性计算机可读存储介质，例如包括计算机程序指令的存储器1932，上述计算机程序指令可由电子设备1900的处理组件1922执行以完成上述方法。
本公开可以是系统、方法和/或计算机程序产品。计算机程序产品可以包括计算机可读存储介质，其上载有用于使处理器实现本公开的各个方面的计算机可读程序指令。
计算机可读存储介质可以是可以保持和存储由指令执行设备使用的指令的有形设备。计算机可读存储介质例如可以是――但不限于――电存储设备、磁存储设备、光存储设备、电磁存储设备、半导体存储设备或者上述的任意合适的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、静态随机存取存储器(SRAM)、便携式压缩盘只读存储器(CD-ROM)、数字多功能盘(DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。这里所使用的计算机可读存储介质不被解释为瞬时信号本身，诸如无线电波或者其他自由传播的电磁波、通过波导或其他传输媒介传播的电磁波(例如，通过光纤电缆的光脉冲)、或者通过电线传输的电信号。
这里所描述的计算机可读程序指令可以从计算机可读存储介质下载到各个计算/处理设备，或者通过网络、例如因特网、局域网、广域网和/或无线网下载到外部计算机或外部存储设备。网络可以包括铜传输电缆、光纤传输、无线传输、路由器、防火墙、交换机、网关计算机和/或边缘服务器。每个计算/处理设备中的网络适配卡或者网络接口从网络接收计算机可读程序指令，并转发该计算机可读程序指令，以供存储在各个计算/处理设备中的计算机可读存储介质中。
用于执行本公开操作的计算机程序指令可以是汇编指令、指令集架构(ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码，所述编程语言包括面向对象的编程语言—诸如Smalltalk、C++等，以及常规的过程式编程语言—诸如“C”语言或类似的编程语言。计算机可读程序指令可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络—包括局域网(LAN)或广域网(WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。在一些实施例中，通过利用计算机可读程序指令的状态信息来个性化定制电子电路，例如可编程逻辑电路、现场可编程门阵列(FPGA)或可编程逻辑阵列(PLA)，该电子电路可以执行计算机可读程序指令，从而实现本公开的各个方面。
这里参照根据本公开实施例的方法、装置(系统)和计算机程序产品的流程图和/或框图描述了本公开的各个方面。应当理解，流程图和/或框图的每个方框以及流程图和/或框图中各方框的组合，都可以由计算机可读程序指令实现。
这些计算机可读程序指令可以提供给通用计算机、专用计算机或其它可编程数据处理装置的处理器，从而生产出一种机器，使得这些指令在通过计算机或其它可编程数据处理装置的处理器执行时，产生了实现流程图和/或框图中的一个或多个方框中规定的功能/动作的装置。也可以把这些计算机可读程序指令存储在计算机可读存储介质中，这些指令使得计算机、可编程数据处理装置和/或其他设 备以特定方式工作，从而，存储有指令的计算机可读介质则包括一个制造品，其包括实现流程图和/或框图中的一个或多个方框中规定的功能/动作的各个方面的指令。
也可以把计算机可读程序指令加载到计算机、其它可编程数据处理装置、或其它设备上，使得在计算机、其它可编程数据处理装置或其它设备上执行一系列操作步骤，以产生计算机实现的过程，从而使得在计算机、其它可编程数据处理装置、或其它设备上执行的指令实现流程图和/或框图中的一个或多个方框中规定的功能/动作。
附图中的流程图和框图显示了根据本公开的多个实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或指令的一部分，所述模块、程序段或指令的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
以上已经描述了本公开的各实施例，上述说明是示例性的，并非穷尽性的，并且也不限于所披露的各实施例。在不偏离所说明的各实施例的范围和精神的情况下，对于本技术领域的普通技术人员来说许多修改和变更都是显而易见的。本文中所用术语的选择，旨在最好地解释各实施例的原理、实际应用或对市场中的技术的改进，或者使本技术领域的其它普通技术人员能理解本文披露的各实施例。

Claims (15)

  1. 一种图像处理方法，其特征在于，包括：
    对待处理图像进行人脸特征提取及人体特征提取，得到所述待处理图像的图像特征，所述图像特征包括人脸特征和/或人体特征，所述待处理图像包括第一图像及第二图像；
    针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果；
    针对未提取到人脸特征的所述第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果；
    根据所述人脸聚类结果及所述人体聚类结果得到针对所述待处理图像的聚类结果。
  2. 根据权利要求1所述的方法，其特征在于，所述人脸聚类结果包括第一结果，所述针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果，包括：
    获取图像数据库中的至少一个已有类别的人脸聚类中心；
    根据所述至少一个已有类别的人脸聚类中心及所述第一图像中提取到的所述人脸特征进行人脸聚类，将所述第一图像聚类至所述已有类别中，得到所述第一图像的所述第一结果。
  3. 根据权利要求2所述的方法，其特征在于，所述人脸聚类结果还包括第二结果，所述针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果，还包括：
    对未聚类到所述已有类别中的所述第一图像进行人脸聚类操作，得到所述第一图像的所述第二结果。
  4. 根据权利要求1至3任一项所述的方法，其特征在于，所述人体聚类结果包括第三结果，所述针对未提取到人脸特征的第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果，包括：
    针对任一所述第二图像，根据提取出人体特征的所述第一图像中的所述人体特征及所述第二图像中的所述人体特征进行人体聚类操作，得到人体聚类子结果；
    根据所述人体聚类子结果确定与所述第二图像归属于同一人体类别的所述第一图像；
    根据所述人脸聚类结果，将所述第二图像添加至与所述第二图像归属于同一人体类别的所述第一图像所属的类别中，得到所述第三结果。
  5. 根据权利要求4所述方法，其特征在于，所述人体聚类结果还包括第四结果，所述针对任一未提取到人脸特征的第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的所述第一图像中的人体特征进行人体聚类操作，得到人体聚类结果，包括：
    获取图像数据库中至少一个已有类别的人体聚类中心；
    对未聚类到人脸类别中的所述第二图像，根据所述第二图像中的所述人体特征及所述至少一个已有类别的所述人体聚类中心进行人体聚类操作，将所述第二图像聚类至所述已有类别中，得到所述第 四结果。
  6. 根据权利要求1至5任一项所述的方法，其特征在于，所述方法还包括：
    根据所述聚类结果将所述待处理图像添加至图像数据库中；
    根据所述待处理图像更新所述图像数据库中至少一个类别的人脸聚类中心及人体聚类中心。
  7. 一种图像处理装置，其特征在于，包括：
    提取模块，用于对待处理图像进行人脸特征提取及人体特征提取，得到所述待处理图像的图像特征，所述图像特征包括人脸特征和/或人体特征，所述待处理图像包括第一图像及第二图像；
    第一聚类模块，用于针对提取到人脸特征的所述第一图像，根据提取到的所述人脸特征进行人脸聚类操作，得到人脸聚类结果；
    第二聚类模块，用于针对未提取到人脸特征的所述第二图像，根据所述人脸聚类结果、所述第二图像中提取的人体特征及提取出人体特征的第一图像中的人体特征进行人体聚类操作，得到人体聚类结果；
    第三聚类模块，用于根据所述人脸聚类结果及所述人体聚类结果得到针对所述待处理图像的聚类结果。
  8. 根据权利要求7所述的装置，其特征在于，所述人脸聚类结果包括第一结果，所述第一聚类模块，还用于：
    获取图像数据库中的至少一个已有类别的人脸聚类中心；
    根据所述至少一个已有类别的人脸聚类中心及所述第一图像中提取到的所述人脸特征进行人脸聚类，将所述第一图像聚类至所述已有类别中，得到所述第一图像的所述第一结果。
  9. 根据权利要求8所述的装置，其特征在于，所述人脸聚类结果还包括第二结果，所述第一聚类模块，还用于：
    对未聚类到所述已有类别中的所述第一图像进行人脸聚类操作，得到所述第一图像的所述第二结果。
  10. 根据权利要求7至9任一项所述的装置，其特征在于，所述人体聚类结果包括第三结果，所述第二聚类模块，还用于：
    针对任一所述第二图像，根据提取出人体特征的所述第一图像中的所述人体特征及所述第二图像中的所述人体特征进行人体聚类操作，得到人体聚类子结果；
    根据所述人体聚类子结果确定与所述第二图像归属于同一人体类别的所述第一图像；
    根据所述人脸聚类结果，将所述第二图像添加至与所述第二图像归属于同一人体类别的所述第一图像所属的类别中，得到所述第三结果。
  11. 根据权利要求10所述的装置，其特征在于，所述人体聚类结果还包括第四结果，所述第二聚类模块，还用于：
    获取图像数据库中至少一个已有类别的人体聚类中心；
    对未聚类到人脸类别中的所述第二图像，根据所述第二图像中的所述人体特征及所述至少一个已有类别的所述人体聚类中心进行人体聚类操作，将所述第二图像聚类至所述已有类别中，得到所述第四结果。
  12. 根据权利要求7至11任一项所述的装置，其特征在于，所述装置还包括：
    添加模块，用于根据所述聚类结果将所述待处理图像添加至图像数据库中；
    更新模块，用于根据所述待处理图像更新所述图像数据库中至少一个类别的人脸聚类中心及人体聚类中心。
  13. 一种电子设备，其特征在于，包括：
    处理器；
    用于存储处理器可执行指令的存储器；
    其中，所述处理器被配置为：执行权利要求1至6中任意一项所述的方法。
  14. 一种计算机可读存储介质，其上存储有计算机程序指令，其特征在于，所述计算机程序指令被处理器执行时实现权利要求1至6中任意一项所述的方法。
  15. 一种计算机程序，包括计算机可读代码，其特征在于，当所述计算机可读代码在电子设备中运行时，所述电子设备中的处理器执行用于实现权利要求1-6中的任一权利要求所述的方法。
PCT/CN2020/093779 2019-08-30 2020-06-01 图像处理方法及装置、电子设备和存储介质 WO2021036382A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
SG11202110569TA SG11202110569TA (en) 2019-08-30 2020-06-01 Image processing method and apparatus, electronic device and storage medium
JP2021552939A JP2022523243A (ja) 2019-08-30 2020-06-01 画像処理方法及び装置、電子機器並びに記憶媒体
US17/488,631 US20220019772A1 (en) 2019-08-30 2021-09-29 Image Processing Method and Device, and Storage Medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910818028.9A CN110569777B (zh) 2019-08-30 2019-08-30 图像处理方法及装置、电子设备和存储介质
CN201910818028.9 2019-08-30

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/488,631 Continuation US20220019772A1 (en) 2019-08-30 2021-09-29 Image Processing Method and Device, and Storage Medium

Publications (2)

Publication Number Publication Date
WO2021036382A1 true WO2021036382A1 (zh) 2021-03-04
WO2021036382A9 WO2021036382A9 (zh) 2022-01-13

Family

ID=68777249

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/093779 WO2021036382A1 (zh) 2019-08-30 2020-06-01 图像处理方法及装置、电子设备和存储介质

Country Status (6)

Country Link
US (1) US20220019772A1 (zh)
JP (1) JP2022523243A (zh)
CN (1) CN110569777B (zh)
SG (1) SG11202110569TA (zh)
TW (1) TW202109360A (zh)
WO (1) WO2021036382A1 (zh)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110569777B (zh) * 2019-08-30 2022-05-06 深圳市商汤科技有限公司 图像处理方法及装置、电子设备和存储介质
CN111783743A (zh) * 2020-07-31 2020-10-16 上海依图网络科技有限公司 一种图像聚类方法及装置
CN112131999B (zh) * 2020-09-17 2023-11-28 浙江商汤科技开发有限公司 身份确定方法及装置、电子设备和存储介质
CN112818867A (zh) * 2021-02-02 2021-05-18 浙江大华技术股份有限公司 人像聚类方法、设备及存储介质
CN112948612B (zh) * 2021-03-16 2024-02-06 杭州海康威视数字技术股份有限公司 人体封面生成方法、装置、电子设备及存储介质
CN113360688B (zh) * 2021-06-28 2024-02-20 北京百度网讯科技有限公司 信息库的构建方法、装置及系统
CN114333039B (zh) * 2022-03-03 2022-07-08 济南博观智能科技有限公司 一种人像聚类的方法、装置及介质
CN115953650B (zh) * 2023-03-01 2023-06-27 杭州海康威视数字技术股份有限公司 特征融合模型的训练方法和装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090129668A1 (en) * 2006-10-14 2009-05-21 Asustek Computer Inc. Image-recognition method and system using the same
CN106874347A (zh) * 2016-12-26 2017-06-20 深圳市深网视界科技有限公司 一种匹配人体特征与mac地址的方法及系统
CN110163096A (zh) * 2019-04-16 2019-08-23 北京奇艺世纪科技有限公司 人物识别方法、装置、电子设备和计算机可读介质
CN110569777A (zh) * 2019-08-30 2019-12-13 深圳市商汤科技有限公司 图像处理方法及装置、电子设备和存储介质

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7734067B2 (en) * 2004-12-07 2010-06-08 Electronics And Telecommunications Research Institute User recognition system and method thereof
US20070237364A1 (en) * 2006-03-31 2007-10-11 Fuji Photo Film Co., Ltd. Method and apparatus for context-aided human identification
JP2010233737A (ja) * 2009-03-30 2010-10-21 Sogo Keibi Hosho Co Ltd 体型判定方法及び体型判定システム並びにそれを用いた監視システム
JP2014229178A (ja) * 2013-05-24 2014-12-08 株式会社東芝 電子機器および表示制御方法、プログラム
JP6427973B2 (ja) * 2014-06-12 2018-11-28 オムロン株式会社 画像認識装置及び画像認識装置への特徴量データ登録方法
CN104731964A (zh) * 2015-04-07 2015-06-24 上海海势信息科技有限公司 基于人脸识别的人脸摘要方法、视频摘要方法及其装置
CN106295469B (zh) * 2015-05-21 2020-04-17 北京文安智能技术股份有限公司 一种基于人脸的来客属性分析方法、装置及系统
CN105893937A (zh) * 2016-03-28 2016-08-24 联想(北京)有限公司 一种图像识别方法及装置
CN106295568B (zh) * 2016-08-11 2019-10-18 上海电力学院 基于表情和行为双模态结合的人类自然状态情感识别方法
CN107644204B (zh) * 2017-09-12 2020-11-10 南京凌深信息科技有限公司 一种用于安防系统的人体识别与跟踪方法
CN107644213A (zh) * 2017-09-26 2018-01-30 司马大大（北京）智能系统有限公司 视频人物提取方法及装置
CN108154171B (zh) * 2017-12-20 2021-04-23 北京奇艺世纪科技有限公司 一种人物识别方法、装置及电子设备
CN108509994B (zh) * 2018-03-30 2022-04-12 百度在线网络技术（北京）有限公司 人物图像聚类方法和装置
CN109213732B (zh) * 2018-06-28 2022-03-18 努比亚技术有限公司 一种改善相册分类的方法、移动终端及计算机可读存储介质
CN109117803B (zh) * 2018-08-21 2021-08-24 腾讯科技（深圳）有限公司 人脸图像的聚类方法、装置、服务器及存储介质
CN109829433B (zh) * 2019-01-31 2021-06-25 北京市商汤科技开发有限公司 人脸图像识别方法、装置、电子设备及存储介质
CN110175555A (zh) * 2019-05-23 2019-08-27 厦门市美亚柏科信息股份有限公司 人脸图像聚类方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090129668A1 (en) * 2006-10-14 2009-05-21 Asustek Computer Inc. Image-recognition method and system using the same
CN106874347A (zh) * 2016-12-26 2017-06-20 深圳市深网视界科技有限公司 一种匹配人体特征与mac地址的方法及系统
CN110163096A (zh) * 2019-04-16 2019-08-23 北京奇艺世纪科技有限公司 人物识别方法、装置、电子设备和计算机可读介质
CN110569777A (zh) * 2019-08-30 2019-12-13 深圳市商汤科技有限公司 图像处理方法及装置、电子设备和存储介质

Also Published As

Publication number Publication date
CN110569777A (zh) 2019-12-13
CN110569777B (zh) 2022-05-06
SG11202110569TA (en) 2021-10-28
JP2022523243A (ja) 2022-04-21
US20220019772A1 (en) 2022-01-20
WO2021036382A9 (zh) 2022-01-13
TW202109360A (zh) 2021-03-01

Similar Documents

Publication Publication Date Title
WO2021036382A1 (zh) 图像处理方法及装置、电子设备和存储介质
WO2021008195A1 (zh) 数据更新方法及装置、电子设备和存储介质
TWI747325B (zh) 目標對象匹配方法及目標對象匹配裝置、電子設備和電腦可讀儲存媒介
WO2020135127A1 (zh) 一种行人识别方法及装置
WO2021196401A1 (zh) 图像重建方法及装置、电子设备和存储介质
WO2020029966A1 (zh) 视频处理方法及装置、电子设备和存储介质
TWI702544B (zh) 圖像處理方法、電子設備和電腦可讀儲存介質
WO2021031645A1 (zh) 图像处理方法及装置、电子设备和存储介质
RU2664003C2 (ru) Способ и устройство для определения ассоциированного пользователя
WO2021093375A1 (zh) 检测同行人的方法及装置、系统、电子设备和存储介质
CN112101238A (zh) 聚类方法及装置、电子设备和存储介质
CN110781957A (zh) 图像处理方法及装置、电子设备和存储介质
CN110659690B (zh) 神经网络的构建方法及装置、电子设备和存储介质
CN110532956B (zh) 图像处理方法及装置、电子设备和存储介质
CN106534951B (zh) 视频分割方法和装置
CN111259967B (zh) 图像分类及神经网络训练方法、装置、设备及存储介质
US20160314164A1 (en) Methods and devices for sharing cloud-based business card
CN109101542B (zh) 图像识别结果输出方法及装置、电子设备和存储介质
WO2016188065A1 (zh) 云名片推荐方法及装置
CN111582383A (zh) 属性识别方法及装置、电子设备和存储介质
TW202145064A (zh) 對象計數方法、電子設備、電腦可讀儲存介質
WO2021164100A1 (zh) 图像处理方法及装置、电子设备和存储介质
WO2023040202A1 (zh) 人脸识别方法及装置、电子设备和存储介质
CN111062407B (zh) 图像处理方法及装置、电子设备和存储介质
CN111625671A (zh) 数据处理方法及装置、电子设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20856671

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021552939

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 01.07.2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20856671

Country of ref document: EP

Kind code of ref document: A1