WO2020151281A9 - 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
WO2020151281A9
WO2020151281A9 PCT/CN2019/113749 CN2019113749W WO2020151281A9 WO 2020151281 A9 WO2020151281 A9 WO 2020151281A9 CN 2019113749 W CN2019113749 W CN 2019113749W WO 2020151281 A9 WO2020151281 A9 WO 2020151281A9
Authority
WO
WIPO (PCT)
Prior art keywords
image
map
feature
binocular
depth
Prior art date
Application number
PCT/CN2019/113749
Other languages
English (en)
French (fr)
Other versions
WO2020151281A1 (zh
Inventor
周尚辰
张佳维
任思捷
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 JP2020573341A priority Critical patent/JP7033674B2/ja
Priority to SG11202106271XA priority patent/SG11202106271XA/en
Priority to KR1020217002881A priority patent/KR20210028218A/ko
Publication of WO2020151281A1 publication Critical patent/WO2020151281A1/zh
Publication of WO2020151281A9 publication Critical patent/WO2020151281A9/zh
Priority to US17/345,042 priority patent/US20210319538A1/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/50Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/045Combinations of networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/0464Convolutional networks [CNN, ConvNet]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/60Image enhancement or restoration using machine learning, e.g. neural networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/73Deblurring; Sharpening
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/50Depth or shape recovery
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/50Depth or shape recovery
    • G06T7/55Depth or shape recovery from multiple images
    • G06T7/593Depth or shape recovery from multiple images from stereo images
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10004Still image; Photographic image
    • G06T2207/10012Stereo images
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10028Range image; Depth image; 3D point clouds
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20084Artificial neural networks [ANN]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20212Image combination
    • G06T2207/20221Image fusion; Image merging

Definitions

  • the present disclosure relates to the field of image processing but is not limited to the field of image processing, and in particular to image processing methods and devices, electronic equipment and storage media for binocular images.
  • binocular vision has been rapidly developed in the fields of smart phones, unmanned driving, drones and robots.
  • Binocular cameras are ubiquitous nowadays, and research on related topics based on binocular images has also been further developed, such as stereo matching, binocular image super-resolution, binocular style conversion and other fields.
  • the image is usually blurred due to factors such as camera shake, out of focus, and high-speed object motion.
  • the optimized method is not satisfactory in terms of performance and efficiency.
  • the embodiments of the present disclosure provide an image processing method and device, electronic equipment, and storage medium that improve the accuracy of binocular images.
  • an image processing method which includes: obtaining a binocular image, wherein the binocular image includes a first image and a second image shot in the same scene for the same object; The first feature map of the binocular image, the first depth map of the binocular image, and the second feature map that combines the image features and depth features of the binocular image; Perform feature fusion processing on the first feature map, the first depth map, and the second feature map of the binocular image to obtain the fusion feature map of the binocular image; perform optimization processing on the fusion feature map of the binocular image to obtain The binocular image after deblurring.
  • an image processing device which includes: an acquisition module configured to acquire binocular images, wherein the binocular images include a first image taken in the same scene for the same object and A second image; a feature extraction module configured to obtain a first feature map of the binocular image, a first depth map of the binocular image, and a second feature that combines image features and depth features of the binocular image Figure; feature fusion module, configured to perform feature fusion processing on the binocular image, the first feature map of the binocular image, the first depth map, and the second feature map to obtain the fusion of the binocular image Feature map; an optimization module configured to perform optimization processing on the fusion feature map of the binocular image to obtain a binocular image after deblurring.
  • an electronic device including: a processor; a memory configured to store executable instructions of the processor; wherein the processor is configured to execute any one of the first aspect The method described.
  • a computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the method described in any one of the first aspects .
  • a computer program product comprising computer program instructions, wherein the computer program instructions are executed by a processor to implement the method in any one of the first aspect .
  • the embodiment of the present disclosure can realize the use of binocular images as input, and perform feature extraction processing on the first image and the second image in the binocular images to obtain the corresponding first feature map, and obtain the first image in the binocular image And the depth map of the second image, and then the obtained features can be fused to obtain a fusion feature containing view information and depth information.
  • the fusion feature contains richer picture information and is more robust to spatial changes.
  • the fusion feature performs optimization processing to obtain a clear binocular image.
  • the embodiment of the present disclosure performs deblurring processing on the binocular image, which improves the accuracy and definition of the image.
  • Fig. 1 shows a flowchart of an image processing method according to an embodiment of the present disclosure
  • Fig. 2 shows a flowchart of step S20 in an image processing method according to an embodiment of the present disclosure
  • Fig. 3 shows a block diagram of a neural network model implementing an image processing method according to an embodiment of the present disclosure
  • Fig. 4 shows a structural block diagram of a context awareness unit according to an embodiment of the present disclosure
  • Fig. 5 shows a flowchart of step S23 in the image processing method according to an embodiment of the present disclosure
  • Fig. 6 shows another flowchart of step S20 in the image processing method according to an embodiment of the present disclosure
  • Fig. 7 shows a flowchart of step S30 in the image processing method according to an embodiment of the present disclosure
  • Figure 8 shows a block diagram of a converged network module according to an embodiment of the present disclosure
  • Fig. 9 shows a flowchart of step S31 in an image processing method according to an embodiment of the present disclosure
  • FIG. 10 shows a block diagram of an image processing device according to an embodiment of the present disclosure
  • FIG. 11 shows a block diagram of an electronic device 800 according to an embodiment of the present disclosure
  • FIG. 12 shows a block diagram of an electronic device 1900 according to an embodiment of the present disclosure.
  • Fig. 1 shows a flowchart of an image processing method according to an embodiment of the present disclosure, wherein the image processing method of the embodiment of the present disclosure can be used to perform deblurring processing on a binocular image to obtain a clear binocular image.
  • the methods of the embodiments of the present disclosure can be applied to binocular cameras, binocular camera equipment, aircraft, or other devices with camera functions, or the embodiments of the present disclosure can also be applied to electronic devices or server devices with image processing, such as mobile phones , Computer equipment, etc.
  • the present disclosure does not specifically limit this, as long as the binocular camera operation can be performed, or the image processing function can be performed, the embodiments of the present disclosure can be applied.
  • the embodiments of the present disclosure will be described below with reference to FIG. 1.
  • the image processing method of the embodiment of the present disclosure may include: S10: Obtain a binocular image, where the binocular image includes a first image and a second image taken in the same scene for the same object.
  • the method of the embodiments of the present disclosure can be applied to a camera device or an image processing device, and binocular images can be acquired by the above-mentioned devices, for example, by the camera device, or transmitted by other devices.
  • the binocular image may include the first image and the second image.
  • the camera device that collects the binocular view may cause blurry images due to various factors (such as device shake, movement of the subject, etc.)
  • the embodiments of the present disclosure can implement defuzzification processing for binocular images, and obtain clear binocular images.
  • the first image and the second image in the binocular image can be configured as a left image and a right image, respectively, or can be configured as an upper side view and a lower side view. It is determined according to the position of the camera lens of the camera device that collects the binocular image, which is not specifically limited in the embodiment of the present disclosure.
  • S20 Obtain a first feature map of the binocular image, a first depth map of the binocular image, and a second feature map that combines image features and depth features of the binocular image.
  • the binocular images may be images of different angles collected on an object at the same time.
  • the depth value of the object can be determined by combining the viewing angle difference of the binocular image.
  • a binocular camera is used to simulate the eyes of a person to capture images of an object from different angles, and two images collected by the camera at the same time can be used as the binocular images. After the binocular image is obtained, the feature map, the depth map, and the feature map combining features and depth information can be extracted from the binocular image.
  • the embodiment of the present disclosure can realize the feature extraction function through a neural network.
  • the neural network can be a convolutional neural network, and the first feature map and the first depth map of the first image and the second image are extracted respectively through the neural network.
  • the neural network may include an image feature extraction module and a depth feature extraction module.
  • the first feature map of the first image and the first feature map of the second image can be obtained respectively, and by The binocular image is input to the depth feature extraction module, and the first depth map of the first image and the first depth map of the second image can be obtained.
  • the image feature of the fused first image and the second feature map of the depth feature can be obtained respectively.
  • the first feature map represents the image features of the first image and the second image, such as the pixel value of each pixel.
  • the first depth map represents the depth features of the first image and the second image, such as the depth information of each pixel.
  • the second feature map combines image features and depth features.
  • the pixels of the first depth map, the first feature map, and the second feature map have a one-to-one correspondence.
  • the structure of the image feature extraction module and the depth feature extraction module are not specifically limited, which may include structures such as convolutional layer, pooling layer, residual module, or fully connected layer. Those skilled in the art can design according to their needs. As long as feature extraction can be achieved, it can be used as an embodiment of the present disclosure. After each feature is obtained, feature fusion processing can be performed to obtain a more accurate feature map based on further fusion of various information.
  • S30 Perform feature fusion processing on the binocular image, the first feature map, the first depth map, and the second feature map of the binocular image to obtain a fusion feature map of the binocular image.
  • the embodiment of the present disclosure can perform feature fusion processing based on the features obtained in step S20, that is, perform feature fusion processing on the original image and the corresponding first feature map, second feature map, and first depth map to obtain the fused feature.
  • Features can contain richer picture information (image features) and are more robust to blurring of spatial changes.
  • the neural network of the embodiment of the present disclosure may include a fusion network module, and the fusion network module may perform the above step S30 by inputting the first feature map, the first depth map, and the second feature map of the first image to the fusion network
  • the module can obtain the fusion feature map of the first image that combines the image information of the first image and the depth information.
  • a fusion feature map of the second image in which the image information of the second image and the depth information are fused can be obtained.
  • a clearer optimized view can be obtained through the obtained fusion feature map.
  • the structure of the fusion feature module is not specifically limited in the embodiments of the present disclosure. It may include structures such as convolutional layer, pooling layer, residual module, or fully connected layer. Those skilled in the art can set according to their needs, as long as The feature fusion can be used as an embodiment of the present disclosure.
  • the fusion can be realized by the feature splicing method after the feature alignment, or it can be realized based on the fusion calculation such as the feature weighted average after the feature alignment.
  • the fusion can be realized by the feature splicing method after the feature alignment, or it can be realized based on the fusion calculation such as the feature weighted average after the feature alignment.
  • S40 Perform optimization processing on the fusion feature map of the binocular image to obtain a binocular image after deblurring processing.
  • the embodiment of the present disclosure can optimize the first fusion feature map and the second fusion feature map through a convolution processing operation, and the effective information in each fusion feature map can be used through the convolution operation to obtain a more accurate optimized view
  • defuzzification of binocular images can be achieved, and the clarity of the view can be increased.
  • the neural network of the embodiment of the present disclosure may further include an optimization module, and the first fusion feature map of the first image and the first fusion feature map of the second image may be input into the optimization module respectively, and the optimization module can be used for at least one volume
  • the product processing operation can fuse and optimize the first fusion feature map of the two images respectively, and the scale of the optimized fusion feature map corresponds to the scale of the original binocular image, and the clarity of the original binocular image is improved .
  • obtaining the first feature map of the binocular image may include:
  • the neural network may include an image feature extraction module (a deblurring network module), and the image feature extraction module may be used to perform step S20 to obtain the first feature map of the binocular image.
  • Fig. 3 shows a block diagram of a neural network model implementing an image processing method according to an embodiment of the present disclosure.
  • the dual images can be respectively input to the image feature extraction module A, the first feature map F L of the first image is obtained according to the first image in the binocular image, and the first feature map of the second image is obtained according to the second image F R.
  • the first convolution processing may be performed on the first image and the second image respectively, and the first convolution processing may use at least one convolution unit to perform corresponding convolution processing.
  • multiple convolution units can be used in sequence to perform the first convolution operation, where the output of the previous convolution unit is used as the input of the next convolution unit.
  • the first middle of the two images can be obtained.
  • Feature maps, where the first intermediate feature maps may respectively include image feature information of corresponding images.
  • the first convolution processing may include standard convolution processing.
  • the standard convolution processing is a convolution operation performed using a convolution kernel or a set convolution step size, and each convolution unit may use a corresponding
  • the convolution kernel performs convolution, or performs convolution according to a preset step size, and finally obtains a first intermediate feature map representing image feature information of the first image and a first intermediate feature map representing image feature information of the second image.
  • the convolution kernel can be a 1*1 convolution kernel or a 3*3 convolution kernel.
  • the convolution kernel used in the embodiment of the present disclosure can be The small convolution kernel can simplify the structure of the neural network while meeting the accuracy requirements of image processing.
  • S22 Perform a second convolution process on the first intermediate feature maps of the first image and the second image respectively to obtain multi-scale second intermediate feature maps corresponding to the first image and the second image respectively;
  • the feature extraction network module in the embodiment of the present disclosure may include a context awareness unit. After obtaining the first intermediate feature map, the first intermediate map can be input into the context awareness unit to obtain second intermediate feature maps of multiple scales.
  • the context awareness unit of the embodiment of the present disclosure may perform a second convolution process on the first intermediate feature map of the first image and the first intermediate feature map of the second image to obtain multiple second intermediate feature maps of different scales.
  • the obtained first intermediate feature map can be input to the context sensing unit, and the context sensing unit of the embodiment of the present disclosure can perform the second convolution processing on the first intermediate feature map.
  • the second intermediate feature map of multiple scales corresponding to the first intermediate feature map can be obtained in a manner that does not require cyclic processing.
  • Fig. 4 shows a structural block diagram of a context awareness unit according to an embodiment of the present disclosure.
  • the first intermediate feature map of the first image and the first intermediate feature map of the second image may be further feature fused and optimized by the context sensing unit, and the second intermediate feature maps of different scales can be obtained at the same time.
  • the second convolution process may be a hole convolution process, where different hole ratios may be used to perform hole convolution on the first intermediate feature map to obtain a second intermediate feature map of a corresponding scale.
  • d is used in FIG. 4 1 , d 2 , d 3, and d 4 perform the second convolution process on the first intermediate feature map with four different first void ratios to obtain 4 second intermediate feature maps of different scales, for example, each second intermediate feature map
  • the scale of can be a 2-fold change relationship, which is not specifically limited in the present disclosure.
  • Those skilled in the art can select different first void ratios to perform the corresponding second convolution according to requirements to obtain the corresponding second intermediate feature map, In addition, the present disclosure does not specifically limit the number of voids.
  • the void rate of dynamic convolution can also be called the dilated rates of void convolution.
  • the void ratio defines the spacing between values when the convolution kernel processes data in the void convolution.
  • the second intermediate feature maps of multiple scales corresponding to the first intermediate feature map of the first image can be respectively obtained, and the second intermediate feature maps of multiple scales corresponding to the first intermediate feature map of the second image can be obtained respectively.
  • the obtained second intermediate feature map may include the feature information of the first intermediate feature map at different scales to facilitate subsequent processing.
  • S23 Perform residual processing on the second intermediate feature maps of each scale of the first image and the second image, respectively, to obtain first feature maps corresponding to the first image and the second image respectively.
  • the second intermediate feature maps of different scales can be further performed by the context sensing unit. Residual error processing obtains a first feature map corresponding to the first image and a first feature map corresponding to the second image.
  • FIG. 5 shows a flowchart of step S23 in the image processing method according to an embodiment of the present disclosure, wherein the residual processing is performed on the second intermediate feature maps of each scale of the first image and the second image to obtain
  • the first feature map corresponding to the first image and the second image respectively includes:
  • S231 Connect the second intermediate feature maps of multiple scales of the first image to obtain a first connection feature map, and respectively connect the second intermediate feature maps of multiple scales of the second image to obtain a second connection feature map.
  • connection processing may be performed on the second intermediate feature maps of each scale of the first image to obtain the first connection feature map, for example, the second intermediate maps are connected in the direction of the channel information.
  • connection processing on the second intermediate feature maps of various scales of the second image to obtain the second connection feature map. For example, to connect each second intermediate image in the direction of the channel information, so as to obtain the first image and The features of the second intermediate feature map of the second image are merged.
  • the convolution unit can be used to perform convolution processing on the first connection feature map and the second connection feature map. This process can further merge the features in each second intermediate feature map, and after the convolution processing
  • the scale of the connected feature map is the same as the scale of the first intermediate feature map.
  • the context awareness unit may also include a convolution unit for feature encoding, where the first connection feature map or the second connection feature map obtained by the connection processing can be input to the convolution unit to perform corresponding convolution.
  • Convolution processing to achieve the feature fusion of the first connection feature map or the second connection feature map, and the first feature map obtained after convolution processing by the convolution unit matches the scale of the first image, and the convolution processing is performed by the convolution unit
  • the subsequent second feature map matches the scale of the second image.
  • the first feature map and the second feature map can respectively reflect the image features of the first image and the second image, such as information such as pixel values of pixels.
  • the convolution unit can be at least one convolution layer, and each convolution layer can use different convolution kernels to perform convolution operations, or the same convolution kernel can also perform convolution operations. Those skilled in the art You can choose by yourself, and the present disclosure does not limit it.
  • S233 Perform addition processing on the first intermediate feature map of the first image and the first connected feature map after convolution processing to obtain a first feature map of the first image, and a comparison of the second image The first intermediate feature map and the second connected feature map after convolution processing are added together to obtain the first feature map of the second image.
  • the first intermediate feature map of the first image and the first connection feature map obtained by the convolution process may be further subjected to addition processing, such as corresponding addition of elements to obtain the first feature map of the first image
  • addition processing such as corresponding addition of elements to obtain the first feature map of the first image
  • the first intermediate feature map of the second image and the second connected feature map after convolution processing are added together to obtain the first feature map of the second image.
  • the embodiment of the present disclosure introduces a multi-branch context awareness unit to avoid While increasing the network model, it obtains rich multi-scale features, and can design a defuzzification neural network through a small convolution kernel, and finally obtains a small and fast binocular defuzzification neural network model.
  • obtaining the first depth map of the first image and the second image may include:
  • the neural network may also include a deep feature extraction module B (as shown in FIG. 3).
  • the depth feature extraction module can obtain the depth information of the first image and the second image, such as the first depth map, which can be embodied in the form of a matrix, and the elements in the matrix can represent the first image or the second image The depth value of the corresponding pixel.
  • the first image and the second image can be combined to form a combined view and then input to the depth extraction module.
  • the image combination method can directly connect the two images in the direction of the upper and lower positions.
  • the two images can also be connected in the left and right direction combination. The present disclosure does not specifically limit this .
  • S202 Perform at least one layer of third convolution processing on the combined view to obtain a first intermediate depth feature map
  • the convolution processing of the combined view can be performed, in which the third convolution processing can be performed at least once, and the third convolution processing can also include at least one convolution unit, wherein each convolution unit
  • the third convolution kernel may be used to perform convolution, or the convolution may be performed according to a third preset step size, and finally the first intermediate depth map representing the depth information of the combined view is obtained.
  • the third convolution kernel can be a 1*1 convolution kernel, or a 3*3 convolution kernel
  • the third preset step size can be 2.
  • the convolution kernel used in the embodiments of the present disclosure may be a small convolution kernel, so that the structure of the neural network can be simplified, and the accuracy requirements of image processing can be met at the same time.
  • S203 Perform a fourth convolution process on the first intermediate depth feature map to obtain second intermediate depth feature maps of multiple scales.
  • the depth extraction module of the embodiment of the present disclosure may also include a context awareness unit for extracting multi-scale features of the first intermediate feature map, that is, after the first intermediate feature map is obtained, the context awareness unit can be used to obtain different scales.
  • the second intermediate depth feature map may also use a different second hole rate to perform the fourth convolution processing of the first intermediate feature map.
  • d 1 , d 2 , d 3 and d 4 are used in FIG. 4
  • Four different second hole ratios perform a second convolution process on the first intermediate depth feature map to obtain four second intermediate depth feature maps of different scales.
  • the scale of each second intermediate depth feature map can be a two-fold change relationship. This disclosure does not specifically limit this.
  • the number of void ratios is not specifically limited in this disclosure.
  • the first void rate and the second void rate in the embodiments of the present disclosure may be the same or different, and the present disclosure does not specifically limit this.
  • the first intermediate depth feature map of the first image and the first intermediate depth feature map of the second image can be respectively input to the context sensing unit, and the context sensing unit can be used to compare each with a different second hole rate.
  • the first intermediate depth feature map performs hole convolution processing to obtain multiple scale second intermediate feature maps corresponding to the first intermediate feature map of the first image, and multiple first intermediate feature maps corresponding to the second image The second intermediate feature map of the scale.
  • S204 Perform residual processing on the second intermediate depth feature and the first intermediate depth map to obtain the first depth map of the first image and the second image respectively, and perform the first convolution processing according to any layer Obtain the second characteristic map.
  • the second intermediate depth feature maps of each scale of the first image can be further connected, such as connecting in the channel direction, and then performing the connection depth map obtained by the connection.
  • Convolution processing the process can further merge the depth features in each second intermediate depth feature map, and the scale of the convolution processing connection depth map is the same as the scale of the first intermediate depth feature map of the first image.
  • the second intermediate depth feature maps of each scale of the second image can be connected, for example, in the channel direction, and then the convolution processing can be performed on the connection depth map obtained by the connection.
  • This process can further merge the various first images. 2. Depth features in the intermediate depth feature map, and the scale of the connected depth map after convolution processing is the same as the scale of the first intermediate depth feature map of the second image.
  • the convolution processed feature map and the corresponding first intermediate depth feature map can be added together, such as element corresponding addition, and then convolution processing is performed on the addition result to obtain the first image and the second image respectively The first depth map.
  • the entire process of the depth extraction module can be realized, and the process of extracting and optimizing the depth information of the first image and the second image can be realized.
  • the embodiment of the present disclosure introduces a multi-branch context awareness unit, which can be used without increasing At the same time as the large network model, it obtains rich multi-scale depth features, which has the characteristics of simple network structure and fast running speed.
  • a second feature map containing the image information and depth information of the first image and the second image can also be obtained.
  • This process can be obtained based on the processing process of the depth extraction module.
  • the third convolution process can be performed at least once in the extraction module, wherein the depth map of the fused image feature can be obtained based on the third convolution process of at least one layer, that is, the second image feature fused with the depth feature of the first image can be obtained A feature map, and a second feature map that combines image features and depth features of the second image.
  • FIG. 7 shows a flowchart of step S30 in the image processing method according to an embodiment of the present disclosure.
  • Performing feature fusion processing on the first feature map, the first depth map, and the second feature map of the binocular image to obtain the fusion feature map of the binocular image may include:
  • S31 Perform calibration processing on the second image according to the first depth map of the first image in the binocular image to obtain the first image mask map, and according to the first depth of the second image in the binocular image The figure performs calibration processing on the first image to obtain a mask image of the second image.
  • the neural network of the embodiment of the present disclosure may also include a fusion network module, which is used to perform the fusion processing of the above-mentioned feature information.
  • FIG. 8 shows a block diagram of the fusion network module according to an embodiment of the present disclosure.
  • the fusion processing result of the first depth map of an image, the first feature map of the first image, and the second feature map of the first image is the fusion feature map of the first image, and according to the second image and the second image
  • the fusion processing result of a depth map, the first feature map of the second image, and the second feature map of the second image obtains the fusion feature map of the second image.
  • the neural network of the present disclosure may further include a feature fusion module C, through which further fusion and optimization of feature information can be performed.
  • the embodiment of the present disclosure can obtain the intermediate feature map of each image of the binocular image according to the calibration map and the mask image corresponding to each image in the binocular image. That is, the calibration map and the mask image of the first image are used to obtain the intermediate fusion feature of the first image, and the calibration map and the mask image of the second image are used to obtain the intermediate fusion feature of the second image.
  • the calibration map refers to the feature map after calibration and processing using depth information.
  • the mask map represents the degree of acceptance of the feature information in the first feature map of the image. The process of obtaining the calibration map and the mask map will be described below.
  • Fig. 9 shows a flowchart of step S31 in the image processing method according to an embodiment of the present disclosure.
  • the calibration process is performed on the second image according to the first depth map of the first image in the binocular image to obtain the first image mask map, and according to the second image of the second image in the binocular image
  • a depth map performs calibration processing on the first image to obtain a mask image of the second image, including:
  • S311 Use the first depth map of the first image in the binocular image to perform alignment processing on the second image to obtain a calibration map of the first image, and use the first depth map of the second image to perform alignment on the first image. The image is aligned to obtain a calibration map of the second image.
  • the depth feature of the first image may be used to perform warp processing of the second image to obtain a calibration map of the first image. And using the depth feature of the second image to perform warp processing of the second image to obtain a calibration map of the second image.
  • the process of performing the alignment processing can be realized by the following formula:
  • the first depth feature baseline*focal length/pixel offset feature
  • the baseline represents the distance between the two lenses of the acquired first image and the second image
  • the focal length refers to the focal length of the two lenses.
  • the first depth map can be determined according to the first depth map of the first image.
  • the first pixel offset feature corresponding to the image, and the second pixel offset feature corresponding to the first depth map is determined according to the first depth map of the second image.
  • the pixel offset feature here refers to the deviation of the pixel value corresponding to the depth feature of each pixel in the first depth map.
  • the embodiment of the present disclosure can use the deviation to align the image, that is, use the first depth of the first image.
  • the first pixel offset feature corresponding to the image is applied to the second image to obtain the calibration map of the first image
  • the second pixel offset feature corresponding to the first depth map of the second image is used to act on the first image to obtain the second image Calibration chart.
  • the second image can be aligned according to the first pixel offset, that is, the pixel characteristics of the second image and the first pixel
  • the offsets are added to obtain the calibration map of the first image.
  • the first image is aligned according to the second pixel offset, that is, the corresponding pixel feature of the first image and the second pixel offset are added to obtain the calibration map of the first image.
  • difference processing can be performed on each image and the corresponding calibration map, and the mask image can be obtained using the result of the difference processing.
  • ⁇ I R is the second difference between the second image and the calibration map of the second image, I R represents the second image, and W R (I L ) represents the calibration map using the second image.
  • the difference between the first image and the calibration map of the first image can be obtained, such as the first difference and the second difference.
  • the first difference and the second difference can be in matrix form respectively. Indicates the deviation of each pixel of the first image and the second image.
  • the difference optimization operation can be performed by the mask network module in the feature fusion module, and the acceptance matrix corresponding to the feature information of the first image and the second image is output, that is, the corresponding mask image.
  • the mask image of the first image may be obtained based on the first difference between the calibration image of the first image and the calibration image of the first image, and the difference between the calibration image based on the second image and the second image may be obtained.
  • the second difference value of the second image is obtained, and the mask image of the first image represents the acceptance degree of the feature information in the first feature image of the first image, and the second image
  • the mask image of represents the degree of acceptance of the feature information in the first feature image of the second image;
  • convolution processing can be performed on the first difference between the first image and its calibration map, such as two convolution processing, and the result of the convolution processing is added to the original first difference. , And then perform convolution processing here and finally output the acceptance degree matrix (mask image) corresponding to each feature information of the first image.
  • the acceptance degree matrix can represent the first feature information of each pixel of the first image The degree of acceptance.
  • the product processing finally outputs a matrix (mask map) of the acceptance degree corresponding to each feature information of the second image, and the matrix of the acceptance degree can represent the acceptance degree of the first feature information of each pixel of the second image.
  • the degree of acceptance can be any value between 0 and 1. According to different designs or training methods of the model, the larger the value, the higher the degree of adoption, or the smaller the value, the higher the degree of adoption. The present disclosure does not specifically limit this.
  • the embodiments of the present disclosure may also use the obtained above-mentioned information, such as the calibration map, the mask map, and the binocular image, to perform feature fusion to obtain an intermediate fusion feature map.
  • the intermediate fusion feature map of the first image may be obtained according to the first preset method, according to the calibration map of the first image and the mask image of the first image, and according to the second preset In this manner, the intermediate fusion feature map of the second image is obtained based on the calibration map of the second image and the mask map of the second image.
  • the expression of the first preset mode is:
  • represents the multiplication of corresponding elements
  • W L (I R ) represents the calibration map obtained after the second image is aligned using the first depth map of the first image
  • M L represents the first image A mask image of an image.
  • the expression of the second preset method is:
  • a second intermediate image is expressed as a fusion characteristics
  • indicates the corresponding elements are multiplied
  • W R (F L) represents the first depth map using the second image view of the alignment process performs a first calibration image is obtained
  • M R represents Mask map of the second image.
  • S33 Obtain a depth feature fusion map of each image of the binocular image according to the first depth map and the second feature map of each image in the binocular image.
  • the embodiment of the present disclosure can also perform a feature fusion process of the first depth map of the two images, where the first depth map of the first image and the second feature map of the first image can be used to obtain the depth feature of the first image
  • the fusion map that is, the second feature map of the first image including the image information and the feature information and the first depth map can be subjected to at least one convolution process to further fuse each depth information and view information to obtain a depth feature fusion map.
  • the first depth map of the second image and the second feature map of the second image may be used to obtain the depth feature fusion map of the second image. That is, the second feature map of the second image including the view information and the feature information and the first depth map may be subjected to at least one convolution process to further fuse each depth information and the view information to obtain a depth feature fusion map.
  • the fusion feature map of the first image can be obtained according to the connection result of the first feature map of the first image, the intermediate fusion feature map of the first image, and the depth feature fusion map of the first image, and according to the The connection result of the first feature map of the second image, the intermediate fusion feature map of the second image, and the depth feature fusion map of the second image obtains the fusion feature map of the second image.
  • the above-mentioned information can be connected, such as in the channel direction, to obtain the fusion feature map of the corresponding view.
  • the fusion feature map obtained by the above method includes the optimized depth information, the view information, and the intermediate fusion feature that combines the depth information and the view information.
  • the convolution processing of the fusion feature map may be further executed to obtain the optimized binocular image corresponding to the binocular image.
  • the performing optimization processing on the fusion feature map of the binocular image to obtain the binocular image after deblurring processing includes:
  • the causes of image blur are very complicated, such as camera shake, out of focus, high-speed object movement, etc.
  • the embodiments of the present disclosure overcome the above technical problems and can be applied to binocular smart phone photography. This method can remove images caused by shaking or rapid motion. Blur, get a clear picture, so that users have a better photo experience.
  • the embodiments of the present disclosure can also be applied to the vision systems of aircraft, robots, or autonomous driving. Not only can it recover the image blur caused by shaking or rapid motion, but the clear pictures obtained can also help other vision systems to perform better. Performance, such as obstacle avoidance system, SLAM reconstruction system, etc.
  • the method of the embodiments of the present disclosure can also be applied to the auxiliary analysis of video surveillance of vehicles.
  • the method can greatly improve the recovery performance of fast motion blur, and can more clearly capture fast-moving vehicle information, such as license plates and driver patterns. Appearance information.
  • the embodiments of the present disclosure can realize the use of binocular images as input, and perform feature extraction processing on the first image and the second image in the binocular images respectively to obtain the corresponding first feature map, and obtain the first image And the depth map of the second image, and then fuse the first feature and depth value of the binocular image to obtain a feature that contains the image information and depth information of the first image and the second image, which contains richer picture information and It is more robust to the blur of spatial changes, and finally the fusion feature is optimized for deblurring to obtain a clear binocular image.
  • 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.
  • 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.
  • FIG. 10 shows a block diagram of an image processing device according to an embodiment of the present disclosure.
  • the image processing device includes: an acquisition module 10 configured to acquire binocular images, wherein the binocular images include The first image and the second image taken in the same scene for the same object; the feature extraction module 20 is configured to obtain the first feature map of the binocular image, the first depth map of the binocular image, and the fusion method The second feature map of the image features and depth features of the binocular image; the feature fusion module 30 is configured to compare the binocular image, the first feature map of the binocular image, the first depth map, and the second The feature map performs feature fusion processing to obtain a fusion feature map of the binocular image; the optimization module 40 is configured to perform optimization processing on the fusion feature map of the binocular image to obtain a binocular image after deblurring processing.
  • the feature extraction module includes an image feature extraction module configured to perform a first convolution process on the first image and the second image, respectively, to obtain the first image and the second image, respectively.
  • Corresponding first intermediate feature map ; perform a second convolution process on the first intermediate feature map of the first image and the second image respectively to obtain the multi-scale corresponding to the first image and the second image A second intermediate feature map; and the second intermediate feature maps of each scale of the first image and the second image are respectively subjected to residual processing to obtain first feature maps corresponding to the first image and the second image respectively.
  • the image feature extraction module is further configured to use a first preset convolution kernel and a first convolution step size to perform convolution processing on the first image and the second image respectively, Obtain first intermediate feature maps corresponding to the first image and the second image respectively.
  • the image feature extraction module is further configured to compare the first intermediate feature map of the first image and the second image according to a plurality of preset different first hole ratios. Perform convolution processing to obtain second intermediate feature maps respectively corresponding to the plurality of first hole rates.
  • the image feature extraction module is further configured to respectively connect second intermediate feature maps of multiple scales of the first image to obtain a first connection feature map, and to respectively connect multiple second image features of the second image.
  • a second intermediate feature map with a scale of two to obtain a second connection feature map ; perform convolution processing on the first connection feature map and the second connection feature map respectively; and perform convolution processing on the first intermediate feature map and convolution of the first image
  • the first connection feature map after the product processing is added to obtain the first feature map of the first image, and the first intermediate feature map of the second image and the second connection feature map after the convolution processing are compared. Add processing to obtain the first feature map of the second image.
  • the feature extraction module further includes a depth feature extraction module configured to combine the first image and the second image to form a combined view; perform at least one layer of the third layer on the combined view Convolution processing to obtain a first intermediate depth feature map; performing a fourth convolution processing on the first intermediate depth feature map to obtain a second intermediate depth feature map of multiple scales; and comparing the second intermediate depth feature with the The first intermediate depth map performs residual processing to obtain the first depth map of the first image and the second image respectively, and the second feature map is obtained according to any layer of third convolution processing.
  • the depth feature extraction module is further configured to perform convolution processing on the combined view at least once by using a second preset convolution kernel and a second convolution step size to obtain the first Middle depth feature map.
  • the depth feature extraction module is further configured to perform convolution processing on the first intermediate depth feature map according to a plurality of preset different second hole ratios, to obtain a multiple The two second void ratios respectively correspond to the second intermediate depth feature maps.
  • the feature fusion module is further configured to perform calibration processing on a second image according to a first depth map of the first image in the binocular image to obtain the first image mask image, And performing calibration processing on the first image according to the first depth map of the second image in the binocular image to obtain the mask image of the second image; based on the calibration map corresponding to each image in the binocular image And the mask map, respectively obtain the intermediate fusion features of each image in the binocular image; obtain the depth of each image of the binocular image according to the first depth map and the second feature map of each image in the binocular image Feature fusion map; and according to the connection results of the first feature map of the first image, the intermediate fusion feature map of the first image, and the depth feature fusion map of the first image of each image in the binocular image, each image is correspondingly obtained The fusion feature map.
  • the feature fusion module is further configured to perform alignment processing on the second image by using the first depth map of the first image in the binocular image to obtain the calibration map of the first image, and use The first depth map of the second image performs alignment processing on the first image to obtain a calibration map of the second image; according to the difference between each image in the binocular image and the corresponding calibration map, all Describe the mask map of the first image and the second image.
  • the fusion feature module is further configured to obtain the first image based on the calibration map of the first image and the mask map of the first image in a first preset manner And obtain the intermediate fusion feature map of the second image based on the calibration map of the second image and the mask map of the second image according to the second preset manner.
  • the expression of the first preset manner is:
  • represents the multiplication of corresponding elements
  • W L (I R ) represents the result of processing the second image using the first depth map of the first image
  • M L represents the first The mask map of the image
  • the expression of the second preset mode is:
  • a second intermediate image is expressed as a fusion characteristics
  • indicates the corresponding elements are multiplied
  • W R (F L) represents a first image depth map using the second execution result of the alignment process of the first image
  • M R denotes a second image Mask image.
  • the optimization module is further configured to perform convolution processing on the fusion feature map of the binocular image to obtain the binocular image after deblurring.
  • 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 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 electronic device can be provided as a terminal, server or other form of device.
  • the embodiment of the present application discloses a computer program product.
  • the computer program product includes computer program instructions, where any of the aforementioned methods is implemented when the computer program instructions are executed by a processor.
  • FIG. 11 shows a block diagram of an electronic device 800 according to an embodiment of the present disclosure.
  • the electrical 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 supply 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 nonvolatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable 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 Programmable Read Only Memory
  • PROM Programmable Read Only Memory
  • ROM Read Only Memory
  • Magnetic Memory Flash Memory
  • Magnetic Disk 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 the generation, management, and distribution of 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).
  • the microphone is configured to receive external audio signals.
  • 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 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 can 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 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 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. 12 shows a block diagram of an electronic device 1900 according to an embodiment of the present disclosure.
  • the electronic device 1900 may be provided as a server. 12
  • 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 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.
  • 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 a transient 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, status 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 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 access the Internet 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.
  • 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 such that when these instructions are 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 flowchart and/or block diagram 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, so that the computer-readable medium storing 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 flowchart and/or block diagram.
  • 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 functions for implementing the specified logical function.
  • Executable instructions may also occur in a different order from the order marked in the drawings. For example, two consecutive blocks can actually be executed 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.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

一种图像处理方法及装置、电子设备和存储介质,其中所述方法包括:获取双目图像,其中,所述双目图像包括针对同一对象在同一场景下拍摄的第一图像和第二图像;获得所述双目图像的第一特征图、所述双目图像的第一深度图,以及融合所述双目图像的图像特征和深度特征的第二特征图;对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图;对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像。

Description

图像处理方法及装置、电子设备和存储介质
相关申请的交叉引用
本申请基于申请号为201910060238.6、申请日为2019年01月22日的中国专利申请提出,并要求该中国专利申请的优先权,该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本公开涉及图像处理领域但不限于图像处理领域,特别涉及双目图像的图像处理方法及装置、电子设备和存储介质。
背景技术
当前双目视觉在智能手机、无人驾驶、无人机和机器人等领域得到了飞速发展。双目相机如今无处不在,且基于双目图像的相关课题研究也得到了进一步的发展,例如在立体匹配、双目图像超分辨、双目风格转换等领域都有所应用。然而,在应用中通常会由于相机晃动、失焦、物体高速运动等因素造成图像模糊的情况。针对该情况,双目去模糊领域只有极少量的研究成果,且优化的方法在性能和效率上都不尽人意。
发明内容
本公开实施例提供了一种提高双目图像精度的图像处理方法及装置、电子设备和存储介质。
根据本公开的一方面,提供了一种图像处理方法,其包括:获取双目图像,其中,所述双目图像包括针对同一对象在同一场景下拍摄的第一图像和第二图像;获得所述双目图像的第一特征图、所述双目图像的第一深度图,以及融合所述双目图像的图像特征和深度特征的第二特征图;对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图;对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像。
根据本公开的第二方面,提供了一种图像处理装置,其包括:获取模块,配置为获取双目图像,其中,所述双目图像包括针对同一对象在同一场景下拍摄的第一图像和第二图像;特征提取模块,配置为获得所述双目图像的第一特征图、所述双目图像的第一深度图,以及融合所述双目图像的图像特征和深度特征的第二特征图;特征融合模块,配置为对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图;优化模块,配置为对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像。
根据本公开的第三方面,提供了一种电子设备,其包括:处理器;配置为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:执行第一方面中任意一项所述的方法。
根据本公开的第四方面,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,其中,所述计算机程序指令被处理器执行时实现第一方面中任意一项所述的方法。
根据本公开的第五方面,提供了一种计算机程序产品,所述计算机程序产品包括计算机程序指令,其中,所述计算机程序指令被处理器执行时实现第一方面中任意一项所述的方法。
本公开实施例可以实现将双目图像作为输入,并分别对双目图像中的第一图像和第二图像执行特征提取处理得到对应的第一特征图,并可以获得双目图像中第一图像和第二图像的深度图,而后可以对获得的特征进行融合,得到包含视图信息和深度信息的融合特征,该融合特征包含更丰富的图片信息且对空间变化的模糊更加鲁棒,最后再将融合特征执行优化处理,得到清晰的双目图像。本公开实施例对双目图像进行去模糊处理, 提高了图像的精度和清晰度。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一种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中步骤S20的流程图;
图3示出根据本公开实施例中实现图像处理方法的神经网络模型的框图;
图4示出根据本公开实施例的上下文感知单元的结构框图;
图5示出根据本公开实施例的图像处理方法中步骤S23的流程图;
图6示出根据本公开实施例的图像处理方法中步骤S20的另一流程图;
图7示出根据本公开实施例的图像处理方法中步骤S30的流程图;
图8示出根据本公开实施例的融合网络模块的框图;
图9示出根据本公开实施例的图像处理方法中步骤S31的流程图;
图10示出根据本公开实施例的图像处理装置的框图;
图11示出根据本公开实施例的一种电子设备800的框图;
图12示出根据本公开实施例的一种电子设备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组件。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、组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一种图像处理方法的流程图,其中本公开实施例的图像处理方法可以用于对双目图像执行去模糊处理,得到清晰的双目图像。本公开实施例的方法可以应用在双目相机、双目摄像设备、飞行器或者其他具有摄像功能的设备中,或者本公开实施例也可以应用在具有图像处理的电子设备或者服务器设备中,如手机、计算机设备等,本公开对此不进行具体限定,只要能够执行双目摄像操作,或者能够执行图像处理功能就可以应用本公开实施例。下面结合图1对本公开实施例进行说明。
如图1所示,本公开实施例的图像处理方法可以包括:S10:获取双目图像,其中,所述双目图像包括针对同一对象在同一场景下拍摄的第一图像和第二图像。
如上所述,本公开实施例的方法可以应用摄像设备或者图像处理设备中,通过上述设备可以获取双目图像,例如通过摄像设备采集,或者通过其他设备传输双目图像。双目图像可以包括第一图像和第二图像,由于在实际应用过程中,采集双目视图的摄像设 备可以会由于各种因素(如设备抖动、拍摄对象的运动等情况),而造成图像模糊或者清晰度较低的情况,本公开实施例可以实现对于双目图像的去模糊化处理,得到清晰的双目图像。其中,根据摄像设备的结构情况的不同,双目图像中第一图像和第二图像可以分别构造为左侧图像和右侧图像,或者,也可以构造为上侧视图和下侧视图,具体可以根据采集双目图像的摄像设备的摄像镜头的位置而确定,本公开实施例对此不进行具体限定。
S20:获得所述双目图像的第一特征图、所述双目图像的第一深度图,以及融合所述双目图像的图像特征和深度特征的第二特征图。
在一些实施例中,所述双目图像可为在一个同一个时刻对一个对象采集的不同角度的图像。如此,结合双目图像的视角差就可以确定出该对象的深度值。例如,利用双目摄像头模拟人的双眼分别对一个对象从不同角度采集图像,在同一个时刻摄像头采集的两个图像就可作为所述双目图像。在得到双目图像之后,即可以提取双目图像中的特征图、深度图,以及融合特征和深度信息的特征图。
本公开实施例可以通过神经网络实现该特征提取的功能,如神经网络可以为卷积神经网络,通过该神经网络分别提取第一图像和第二图像的第一特征图和第一深度图。神经网络可以包括图像特征提取模块和深度特征提取模块,通过将双目图像输入至图像特征提取模块,可以分别获得第一图像的第一特征图以及第二图像的第一特征图,以及通过将双目图像输入至深度特征提取模块,可以获得第一图像的第一深度图以及第二图像的第一深度图,同时还可以分别获取融合第一图像的图像特征和深度特征第二特征图,以及融合第二图像的图像特征和深度特征的第二特征图。第一特征图表示第一图像和第二图像的图像特征,如各像素点的像素值等信息。第一深度图表示第一图像和第二图像的深度特征,如各像素点的深度信息。第二特征图中融合了图像特征和深度特征。并且,第一深度图、第一特征图以及第二特征图的各像素点一一对应。
图像特征提取模块和深度特征提取模块的结构本公开实施例不作具体限定,其中可以包括例如卷积层、池化层、残差模块或者全连接层等结构,本领域技术人员可以根据需求进行设定,只要能够实现特征提取即可以作为本公开实施例。在获得各特征之后,则可以执行特征融合处理,进一步融合各信息的基础上得到更精确的特征图。
S30:对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图。
本公开实施例可以根据步骤S20得到的各特征,执行特征融合处理,即可以对原始图像以及对应的第一特征图、第二特征图和第一深度图执行特征融合处理得到融合特征,该融合特征中可以包含更丰富的图片信息(图像特征)且对空间变化的模糊更加鲁棒。
例如,本公开实施例的神经网络可以包括融合网络模块,该融合网络模块可以执行上述步骤S30,通过将第一图像的第一特征图、第一深度图以及第二特征图输入至该融合网络模块,可以得到融合了第一图像的图像信息和深度信息的第一图像的融合特征图。对应的,将第二图像的第一特征图、第一深度图以及第二特征图输入至融合网络模块,可以得到融合了第二图像的图像信息和深度信息的第二图像的融合特征图。通过得到的融合特征图能够得到更为清晰的优化视图。其中,融合特征模块的结构本公开实施例也不作具体限定,其中可以包括例如卷积层、池化层、残差模块或者全连接层等结构,本领域技术人员可以根据需求进行设定,只要能够实现特征融合即可以作为本公开实施例。
在进行特征图和深度图的融合时,可以通过特征对齐之后的特征拼接方式来实现融合,也可以是基于特征对齐之后的特征加权平均等融合计算实现特征融合。特征融合的方式有很多种,在此处就不再做进一步的限定。
S40:对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像。其中,本公开实施例可以通过卷积处理操作对第一融合特征图和第二融合特征图进行优 化,通过卷积操作可以利用各融合特征图中的有效信息,得到精确度更高的优化视图,通过本公开实施例可以实现双目图像的去模糊化,增加视图的清晰度。
其中,本公开实施例的神经网络还可以包括优化模块,第一图像的第一融合特征图和第二图像的第一融合特征图可以分别被输入至优化模块中,通过优化模块的至少一次卷积处理操作,可以分别对两个图像的第一融合特征图进行融合和优化,得到优化后的融合特征图的尺度与原始的双目图像的尺度对应,并提高了原始双目图像的清晰度。
下面分别对各过程进行详细说明。如上述所述,在获得双目图像之后可以分别对双目图像中的第一图像和第二图像执行特征提取处理。图2示出根据本公开实施例的图像处理方法中步骤S20的流程图。其中,获得所述双目图像的第一特征图,可以包括:
S21:对所述第一图像和第二图像分别执行第一卷积处理,得到所述第一图像和第二图像分别对应的第一中间特征图。本公开实施例中,神经网络可以包括图像特征提取模块(去模糊网络模块),可以利用该图像特征提取模块执行步骤S20,得到双目图像的第一特征图。图3示出根据本公开实施例中实现图像处理方法的神经网络模型的框图。其中,可以将双图像分别输入至图像特征提取模块A中,根据双目图像中第一图像得到第一图像的第一特征图F L,以及根据第二图像得到第二图像的第一特征图F R
其中,首先可以对第一图像和第二图像分别执行第一卷积处理,该第一卷积处理可以利用至少一个卷积单元执行相应的卷积处理。例如可以依次利用多个卷积单元执行该第一卷积操作,其中前一个卷积单元的输出作为下一个卷积单元的输入,通过第一卷积处理,可以得到两个图像的第一中间特征图,其中第一中间特征图可以分别包括对应图像的图像特征信息。在本实施例中,第一卷积处理可以包括标准卷积处理,标准卷积处理为利用卷积核或者具有设定卷积步长执行的卷积操作,各卷积单元可以为利用相应的卷积核执行卷积,或者按照预设步长执行卷积,最终得到表征第一图像的图像特征信息的第一中间特征图以及表征第二图像的图像特征信息的第一中间特征图。其中,卷积核可以为1*1的卷积核,也可以为3*3的卷积核,本领域技术人员可以根据需求进行选择和设定,本公开实施例采用的卷积核可以为小卷积核,从而可以简化神经网络的结构,同时满足图像处理的精度需求。
S22:对所述第一图像和第二图像的所述第一中间特征图分别执行第二卷积处理,得到所述第一图像和第二图像分别对应的多尺度的第二中间特征图;
本公开实施例中的特征提取网络模块中可以包括上下文感知单元,在获得第一中间特征图后,可以将第一中间图输入至上下文感知单元中,得到多个尺度的第二中间特征图。
本公开实施例的上下文感知单元可以对第一图像的第一中间特征图以及第二图像的第一中间特征图执行第二卷积处理,得到多个不同尺度的第二中间特征图。
即,在执行第一卷积处理之后,可以将获得的第一中间特征图输入至上下文感知单元,本公开实施例的上下文感知单元可以对第一中间特征图进行第二卷积处理,该过程可以不需要循环处理的方式即可以得到与第一中间特征图对应的多个尺度的第二中间特征图。
图4示出根据本公开实施例的上下文感知单元的结构框图。其中,可以通过上下文感知单元分别对第一图像的第一中间特征图和第二图像的第一中间特征图进行进一步的特征融合和优化处理,并同时得到不同尺度的第二中间特征图。
其中,第二卷积处理可以为空洞卷积处理,其中可以采用不同的空洞率分别对第一中间特征图执行空洞卷积,得到相应尺度的第二中间特征图,例如,图4中采用d 1、d 2、d 3以及d 4四个不同的第一空洞率对第一中间特征图执行第二卷积处理,得到4个不同尺度的第二中间特征图,例如各第二中间特征图的尺度可以为2倍变化的关系,本公开对此不进行具体限定,本领域技术人员可以根据需求选择不同的第一空洞率执行对应的第二卷 积,得到相应的第二中间特征图,另外,对于空洞率的数量本公开也不作具体限定。空动卷积的空洞率又可以称之为空洞卷积的扩张率(dilated rates)。空洞率定义了空洞卷积中卷积核处理数据时各值的间距。
根据上述过程,即可以分别得到第一图像的第一中间特征图分别对应的多个尺度的第二中间特征图,以及得到第二图像的第一中间特征图分别对应的多个尺度的第二中间特征图。得到的第二中间特征图可以包括第一中间特征图在不同尺度下的特征信息,方便后续的处理过程。
S23:对所述第一图像和第二图像的各尺度的第二中间特征图分别执行残差处理,得到所述第一图像和第二图像分别对应的第一特征图。
在得到对应于第一图像的不同尺度的第二中间特征图,以及对应于第二图像的不同尺度的第二特征图之后,可以进一步通过上下文感知单元分别对不同尺度的第二中间特征图进行残差处理,得到对应于第一图像的第一特征图,以及对应于第二图像的第一特征图。
图5示出根据本公开实施例的图像处理方法中步骤S23的流程图,其中,所述对所述第一图像和第二图像的各尺度的第二中间特征图执行残差处理,得到所述第一图像和第二图像分别对应的第一特征图(步骤S23),包括:
S231:分别连接所述第一图像的多个尺度的第二中间特征图得到第一连接特征图,以及分别连接第二图像的多个尺度的第二中间特征图得到第二连接特征图。
本公开实施例在对第一中间特征图执行多尺度处理之后,还可以对获得的多个尺度的第二中间特征图执行连接处理,继而得到对应的包括不同尺度信息的特征图。
在一些实施例中,可以分别对第一图像的各个尺度的第二中间特征图执行连接处理,得到第一连接特征图,例如对各个第二中间图在通道信息的方向上进行连接。同时还可以对第二图像的各个尺度的第二中间特征图执行连接处理得到第二连接特征图,例如对各个第二中间图在通道信息的方向上进行连接,从而可以得到针对第一图像和第二图像的第二中间特征图的特征进行融合。
S232:分别对所述第一连接特征图和第二连接特征图执行卷积处理。
基于步骤S231的处理结果,可以分别利用卷积单元对第一连接特征图和第二连接特征图执行卷积处理,该过程可以进一步融合各个第二中间特征图内的特征,并且卷积处理后的连接特征图的尺度与第一中间特征图的尺度相同。
在一些实施例中,上下文感知单元中还可以包括卷积单元,用于特征编码,其中可以将连接处理得到的第一连接特征图或者第二连接特征图输入至该卷积单元执行相应的卷积处理,实现第一连接特征图或者第二连接特征图的特征融合,同时通过该卷积单元卷积处理后得到的第一特征图与第一图像的尺度匹配,通过卷积单元卷积处理后的第二特征图与第二图像的尺度匹配。第一特征图和第二特征图分别能够体现第一图像和第二图像的图像特征,如像素点的像素值等信息。
其中,该卷积单元可以至少一层的卷积层,每层卷积层可以利用不同的卷积核执行卷积操作,或者也可以利用相同的卷积核执行卷积操作,本领域技术人员可以自行选择,本公开对此不作限定。
S233:对所述第一图像的第一中间特征图和卷积处理后的第一连接特征图执行相加处理,得到所述第一图像的第一特征图,以及对所述第二图像的第一中间特征图和卷积处理后的第二连接特征图执行相加处理,得到所述第二图像的第一特征图。
基于步骤S232的处理结果,可以进一步将第一图像的第一中间特征图和卷积处理得到的第一连接特征图进行相加处理,如元素对应相加,得到第一图像的第一特征图,对应的,将第二图像的第一中间特征图和卷积处理后的第二连接特征图进行相加处理,得到第二图像的第一特征图。
通过上述配置,即可以实现去模糊网络模块的全过程,可以实现第一图像和第二图像的特征信息的优化和提取的过程,本公开实施例通过引入多分支的上下文感知单元,可以在不增大网络模型的同时,获取丰富的多尺度特征,且可以通过小卷积核设计去模糊神经网络,最终得到一个空间占用小且快速的双目去模糊的神经网络模型。
另外,步骤S20中还可以获得第一图像和第二图像的第一深度图。图6示出根据本公开实施例的图像处理方法中步骤S20的另一流程图。其中,获取第一图像和第二图像的第一深度图,可以包括:
S201:将所述第一图像和第二图像进行组合,形成组合视图。
本公开实施例中,神经网络还可以包括深度特征提取模块B(如图3所示)。通过该深度特征提取模块可以获得第一图像和第二图像的深度信息,如第一深度图,该第一深度图可以以矩阵的形式体现,矩阵中的元素可以表示第一图像或者第二图像对应像素点的深度值。
首先,可以将第一图像和第二图像组合,形成组合视图后输入至深度提取模块。其中,图像组合的方式可以直接将两个图像以上下位置的方向连接到一起,在其他的实施例中,也可以采用左右方向组合的方式连接该两个图像,本公开对此不进行具体限定。
S202:对所述组合视图执行至少一层第三卷积处理得到第一中间深度特征图;
在得到组合视图之后,即可以执行该组合视图的卷积处理,其中可以执行至少一次第三卷积处理,同样的该第三卷积处理也可以包括至少一个卷积单元,其中各卷积单元可以为利用第三卷积核执行卷积,或者按照第三预设步长执行卷积,最终得到表征组合视图的深度信息的第一中间深度图。其中,第三卷积核可以为1*1的卷积核,也可以为3*3的卷积核,第三预设步长可以为2,本领域技术人员可以根据需求进行选择和设定,本公开实施例对此不进行限定。其中本公开实施例采用的卷积核可以为小卷积核,从而可以简化神经网络的结构,同时满足图像处理的精度需求。
S203:对所述第一中间深度特征图执行第四卷积处理,得到多个尺度的第二中间深度特征图。
进一步地,本公开实施例的深度提取模块中也可以包括上下文感知单元,用于提取第一中间特征图的多尺度特征,即在得到第一中间特征图后,可以采用上下文感知单元得到不同尺度的第二中间深度特征图。其中,深度提取模块中的上下文感知单元,也可以采用不同的第二空洞率执行第一中间特征图的第四卷积处理,例如,图4中采用d 1、d 2、d 3以及d 4四个不同的第二空洞率对第一中间深度特征图执行第二卷积处理,得到4个不同尺度的第二中间深度特征图。例如各第二中间深度特征图的尺度可以为2倍变化的关系,本公开对此不进行具体限定,本领域技术人员可以根据需求选择不同的空洞率执行对应的第四卷积处理,得到相应的第二中间深度特征图,另外,对于空洞率的数量本公开也不作具体限定。本公开实施例的第一空洞率和第二空洞率可以相同,也可以不同,本公开对此不进行具体限定。
即在步骤S203中,可以分别将第一图像的第一中间深度特征图和第二图像的第一中间深度特征图输入至上下文感知单元,并利用上下文感知单元通过不同的第二空洞率对各第一中间深度特征图执行空洞卷积处理,得到与第一图像的第一中间特征图对应的多个尺度的第二中间特征图,以及与第二图像的第一中间特征图对应的多个尺度的第二中间特征图。
S204:对所述第二中间深度特征与所述第一中间深度图执行残差处理,分别得到所述第一图像和第二图像的第一深度图,以及根据任意一层第一卷积处理获得所述第二特征图。
本公开实施例中,基于步骤S203的处理结果,可以进一步将第一图像的各尺度的第二中间深度特征图进行连接,如在通道方向上进行连接,而后对连接得到的连接深度图 他执行卷积处理,该过程可以进一步融合各个第二中间深度特征图内的深度特征,并且卷积处理后的连接深度图的尺度与第一图像的第一中间深度特征图的尺度相同。对应的,可以将第二图像的各尺度的第二中间深度特征图进行连接,如在通道方向上进行连接,而后对连接得到的连接深度图他执行卷积处理,该过程可以进一步融合各个第二中间深度特征图内的深度特征,并且卷积处理后的连接深度图的尺度与第二图像的第一中间深度特征图的尺度相同。
而后,可以将卷积处理后的特征图和对应的第一中间深度特征图进行相加处理,如元素对应相加,而后对相加结果执行卷积处理,分别得到第一图像和第二图像的第一深度图。
通过上述配置,即可以实现深度提取模块的全过程,可以实现第一图像和第二图像的深度信息的提取和优化的过程,本公开实施例通过引入多分支的上下文感知单元,可以在不增大网络模型的同时,获取丰富的多尺度深度特征,具有网络结构简单且运行速度快的特点。
在此需要说明的是,步骤S20中还可以获得包含所述第一图像和第二图像的图像信息和深度信息的第二特征图,该过程可以基于深度提取模块的处理过程获得,由于在深度提取模块中可以执行至少一次的第三卷积处理,其中可以基于至少一层的第三卷积处理得到融合图像特征的深度图,即可以获取融合第一图像的图像特征和深度特征的第二特征图,以及融合第二图像的图像特征和深度特征的第二特征图。
在执行步骤S20之后,可以对得到的各特征执行特征融合处理,图7示出根据本公开实施例的图像处理方法中步骤S30的流程图,其中,,所述对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图(步骤S30),可以包括:
S31:根据所述双目图像中第一图像的第一深度图对第二图像执行校准处理,获得所述第一图像掩模图,以及根据所述双目图像中第二图像的第一深度图对第一图像执行校准处理,获得所述第二图像的掩模图。
本公开实施例的神经网络还可以包括融合网络模块,其用于执行上述特征信息的融合处理,图8示出根据本公开实施例的融合网络模块的框图,其中,可以根据第一图像、第一图像的第一深度图、第一图像的第一特征图以及第一图像的第二特征图的融合处理结果,得到第一图像的融合特征图,以及根据第二图像、第二图像的第一深度图、第二图像的第一特征图以及第二图像的第二特征图的融合处理结果,得到第二图像的融合特征图。
在一些实施例中,如上所述,本公开的神经网络还可以包括特征融合模块C,通过该特征融合模块C可以执行特征信息的进一步融合和优化。
首先,本公开实施例可以根据双目图像中各图像对应的校准图和掩模图,得到双目图像各图像的中间特征图。即利用第一图像的校准图和掩模图得到第一图像的中间融合特征,以及利用第二图像的校准图和掩模图得到第二图像的中间融合特征。其中校准图是指利用深度信息校准处理后的特征图。掩模图表示图像的第一特征图中特征信息的被采纳度。下面对校准图和掩模图的获取过程进行说明。
图9示出根据本公开实施例的图像处理方法中步骤S31的流程图。其中,所述根据所述双目图像中第一图像的第一深度图对第二图像执行校准处理,获得所述第一图像掩模图,以及根据所述双目图像中第二图像的第一深度图对第一图像执行校准处理,获得所述第二图像的掩模图,包括:
S311:利用双目图像中第一图像的第一深度图对第二图像执行对齐处理,得到所述第一图像的校准图,以及利用所述第二图像的第一深度图对所述第一图像执行对齐处理,得到所述第二图像的校准图。
本公开实施例,可以利用第一图像的深度特征执行第二图像的对齐(warp)处理,得到第一图像的校准图。以及利用第二图像的深度特征执行第二图像的对齐(warp)处理,得到第二图像的校准图。
其中,执行对齐处理的过程可以通过下式实现:
第一深度特征=基线*焦距/像素偏移特征;
其中,基线表示获取的第一图像和第二图像的两个镜头之间的距离,焦距是指两个镜头的焦距,通过上述方式可以根据第一图像的第一深度图确定与该第一深度图对应的第一像素偏移特征,以及根据第二图像的第一深度图确定与该第一深度图对应的第二像素偏移特征。这里的像素偏移特征是指与第一深度图中各像素点的深度特征对应的像素值的偏差,本公开实施例可以利用该偏差对图像进行对齐处理,即利用第一图像的第一深度图对应的第一像素偏移特征作用于第二图像,得到第一图像的校准图,利用第二图像的第一深度图对应的第二像素偏移特征作用与第一图像,得到第二图像的校准图。
其中,在得到第一图像的第一深度图对应的第一像素偏移量之后,可以将第二图像按照该第一像素偏移量执行对齐处理,即将第二图像的像素特征与第一像素偏移量相加,得到第一图像的校准图。以及将第一图像按照该第二像素偏移量执行对齐处理,即将第一图像的对应像素特征与第二像素偏移量相加,得到第一图像的校准图。
S312:根据双目图像中各图像与对应的校准图之间的差异,分别得到所述第一图像和第二图像的掩模图。
在得到每个图像的校准图之后,可以将各图像与对应的校准图执行差值处理,并利用该差值处理的结果得到掩模图。
其中,第一图像与第一图像的校准图之间的差值可以表示为△I L=|I L-W L(I R)|,第二图像与第二图像的校准图之间的差值可以表示为△I R=|I R-W R(I L)|,其中,△I L为第一图像与第一图像的校准图之间的第一差值的校准图,I L表示第一图像,W L(I R)表示利用第一图像的第一深度图执行第二图像的对齐处理后得到的校准图。△I R第二图像与第二图像的校准图之间的第二差值,I R表示第二图像,W R(I L)表示利用第二图像的校准图。
通过上述过程,可以得到第一图像与第一图像的校准图之间的差值,如第一差值和第二差值,该第一差值和第二差值可以分别为矩阵形式,可以表示第一图像和第二图像各像素点的偏差。此时可以通过特征融合模块中的掩模网络模块执行该差值的优化操作,并输出对应于第一图像和第二图像的特征信息的被采纳度矩阵,即对应的掩模图。
其中,可以基于所述第一图像和第一图像的校准图的之间的第一差值,获得第一图像的掩模图,以及基于所述第二图像和第二图像的校准图之间的第二差值,获得第二图像的掩模图,所述第一图像的掩模图表示所述第一图像的第一特征图中的特征信息的被采纳度,以及所述第二图像的掩模图表示第二图像的第一特征图中的特征信息的被采纳度;
如图8所示,可以对第一图像及其校准图之间的第一差值执行卷积处理,如两次卷积处理,并将卷积处理后的结果与原始第一差值相加,而后在此进行卷积处理最终输出与第一图像的各特征信息对应的被采纳程度的矩阵(掩模图),该被采纳程度的矩阵可以表示第一图像各像素点的第一特征信息的被采纳度。另外,可以对第二图像及其校准图之间的第二差值执行卷积处理,如两次卷积处理,并将卷积处理后的结果与原始差值相加,而后在此进行卷积处理最终输出与第二图像的各特征信息对应的被采纳程度的矩阵(掩 模图),该被采纳程度的矩阵可以表示第二图像各像素点的第一特征信息的被采纳度。该被采纳度可以为0到1之间的任意数值,按照不同的设计或者模型的训练方式,可以是该数值越大被采纳度越高,也可以是数值越小,被采纳度越高,本公开对此不进行具体限定。
S32:基于所述双目图像中各图像对应的所述校准图和掩模图,分别获得所述双目图像中各图像的中间融合特征。
本公开实施例还可以利用得到的上述信息,如校准图、掩模图以及双目图像,进行特征融合,得到中间融合特征图。
在一些实施例中,可以按照第一预设方式,根据第一图像的校准图,以及所述第一图像的掩模图得到所述第一图像的中间融合特征图,并按照第二预设方式,基于所述第二图像的校准图,以及所述第二图像的掩模图得到所述第二图像的中间融合特征图。其中,第一预设方式的表达式为:
Figure PCTCN2019113749-appb-000001
其中,
Figure PCTCN2019113749-appb-000002
表示为第一图像的中间融合特征,⊙表示对应元素相乘,W L(I R)表示利用第一图像的第一深度图执行第二图像的对齐处理后得到的校准图,M L表示第一图像的掩模图。
第二预设方式的表达式为:
Figure PCTCN2019113749-appb-000003
其中,
Figure PCTCN2019113749-appb-000004
表示为第二图像的中间融合特征,⊙表示对应元素相乘,W R(F L)表示利用第二图像的第一深度图执行第一图像的对齐处理后得到的校准图,M R表示第二图像的掩模图。
S33:根据所述双目图像中各图像的第一深度图和第二特征图,获得所述双目图像各图像的深度特征融合图。
进一步的,本公开实施例还可以执行两个图像的第一深度图的特征融合过程,其中可以将第一图像的第一深度图以及第一图像的第二特征图得到第一图像的深度特征融合图,即可以将包括了图像信息和特征信息的第一图像的第二特征图与第一深度图执行至少一次卷积处理,进一步融合各深度信息和视图信息,得到深度特征融合图。
对应的,可以利用所述第二图像的第一深度图以及第二图像的第二特征图得到第二图像的深度特征融合图。即可以将包括了视图信息和特征信息的第二图像的第二特征图与第一深度图执行至少一次卷积处理,进一步融合各深度信息和视图信息,得到深度特征融合图。
S34:根据所述双目图像中各图像的第一图像的第一特征图、第一图像的中间融合特征图以及第一图像的深度特征融合图的连接结果,对应的得到各图像的所述融合特征图。
其中,可以根据所述第一图像的第一特征图、第一图像的中间融合特征图以及第一图像的深度特征融合图的连接结果得到所述第一图像的融合特征图,以及根据所述第二图像的第一特征图、第二图像的中间融合特征图以及第二图像的深度特征融合图的连接结果得到所述第二图像的融合特征图。
在本公开实施例中,在得到各第一特征图中间融合特征图以及深度特征融合图之后,可以将上述信息连接,如在通道方向上进行连接,得到相应视图的融合特征图。
通过上述方式得到的融合特征图中包括了优化处理后的深度信息、视图信息,以及 融合有深度信息和视图信息的中间融合特征。对应的步骤S40中,可以进一步执行融合特征图的卷积处理,得到与双目图像的对应的优化后的双目图像。其中,所述对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像,包括:
对所述第一图像的融合特征图执行卷积处理,得到所述优化的第一图像,以及对所述第二图像的融合特征图执行卷积处理,得到所述优化的第二图像。
通过S40,一方面可以得到与原始双目图像尺度匹配的优化图像,另一方面可以更加深入的融合各特征,并提高信息的精度。
由于图像模糊产生的原因非常复杂,比如:相机晃动、失焦、物体高速运动等。而现有的图像编辑工具很难复原这种复杂的模糊图像,本公开实施例克服了上述技术问题,并可以应用在双目智能手机摄像,利用该方法可以去除由抖动或快速运动产生的图像模糊,得到清晰的图片,使用户有更好的拍照体验。另外,本公开实施例还可以应用在飞行器、机器人或自动驾驶的视觉系统上,不仅可以恢复因抖动或快速运动产生的图像模糊,得到的清晰的图片还有助于其他视觉系统发挥更好的性能,如避障系统、SLAM重建系统等。
本公开实施例的方法还可以应用在车辆的视频监控辅助分析中,该方法对快速运动模糊的复原性能有大幅度的提高,可以更清晰地捕捉快速运动的车辆信息,如车牌和驾驶员样貌信息。
综上所述,本公开实施例可以实现将双目图像作为输入,可以分别双目图像中的第一图像和第二图像执行特征提取处理得到对应的第一特征图,并可以获得第一图像和第二图像的深度图,然后对双目图像的第一特征和深度值进行融合,得到包含第一图像和第二图像的图像信息和深度信息的特征,该特征包含更丰富的图片信息且对空间变化的模糊更加鲁棒,最后再将融合特征执行去模糊处理的优化处理,得到清晰的双目图像。
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。
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。
图10示出根据本公开实施例的一种图像处理装置的框图,如图10所示,所述图像处理装置包括:获取模块10,配置为获取双目图像,其中,所述双目图像包括针对同一对象在同一场景下拍摄的第一图像和第二图像;特征提取模块20,配置为获得所述双目图像的第一特征图、所述双目图像的第一深度图,以及融合所述双目图像的图像特征和深度特征的第二特征图;特征融合模块30,配置为对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图;优化模块40,配置为对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像。
在一些可能的实施方式中,所述特征提取模块包括图像特征提取模块,配置为对所述第一图像和第二图像分别执行第一卷积处理,得到所述第一图像和第二图像分别对应的第一中间特征图;对所述第一图像和第二图像的所述第一中间特征图分别执行第二卷积处理,得到所述第一图像和第二图像分别对应的多尺度的第二中间特征图;以及对所述第一图像和第二图像的各尺度的第二中间特征图分别执行残差处理,得到所述第一图像和第二图像分别对应的第一特征图。
在一些可能的实施方式中,所述图像特征提取模块,还配置为利用第一预设卷积核 以及第一卷积步长分别对所述第一图像和第二图像分别执行卷积处理,得到所述第一图像和第二图像分别对应的第一中间特征图。
在一些可能的实施方式中,所述图像特征提取模块,还配置为分别按照预设的多个不同的第一空洞率,对所述第一图像和第二图像的所述第一中间特征图执行卷积处理,得到与该多个第一空洞率分别对应的第二中间特征图。
在一些可能的实施方式中,所述图像特征提取模块,还配置为分别连接所述第一图像的多个尺度的第二中间特征图得到第一连接特征图,以及分别连接第二图像的多个尺度的第二中间特征图得到第二连接特征图;分别对所述第一连接特征图和第二连接特征图执行卷积处理;以及对所述第一图像的第一中间特征图和卷积处理后的第一连接特征图执行相加处理,得到第一图像的第一特征图,以及对所述第二图像的第一中间特征图和卷积处理后的第二连接特征图执行相加处理,得到所述第二图像的第一特征图。
在一些可能的实施方式中,所述特征提取模块还包括深度特征提取模块,配置为将所述第一图像和第二图像进行组合,形成组合视图;对所述组合视图执行至少一层第三卷积处理得到第一中间深度特征图;对所述第一中间深度特征图执行第四卷积处理,得到多个尺度的第二中间深度特征图;以及对所述第二中间深度特征与所述第一中间深度图执行残差处理,分别得到所述第一图像和第二图像的第一深度图,以及根据任意一层第三卷积处理获得所述第二特征图。
在一些可能的实施方式中,所述深度特征提取模块,还配置为利用第二预设卷积核以及第二卷积步长对所述组合视图执行至少一次卷积处理,得到所述第一中间深度特征图。
在一些可能的实施方式中,所述深度特征提取模块,还配置为分别按照预设的多个不同的第二空洞率,对所述第一中间深度特征图执行卷积处理,得到与该多个第二空洞率分别对应的第二中间深度特征图。
在一些可能的实施方式中,所述特征融合模块,还配置为根据所述双目图像中第一图像的第一深度图对第二图像执行校准处理,获得所述第一图像掩模图,以及根据所述双目图像中第二图像的第一深度图对第一图像执行校准处理,获得所述第二图像的掩模图;基于所述双目图像中各图像对应的所述校准图和掩模图,分别获得所述双目图像中各图像的中间融合特征;根据所述双目图像中各图像的第一深度图和第二特征图,获得所述双目图像各图像的深度特征融合图;以及根据所述双目图像中各图像的第一图像的第一特征图、第一图像的中间融合特征图以及第一图像的深度特征融合图的连接结果,对应的得到各图像的所述融合特征图。
在一些可能的实施方式中,所述特征融合模块,还配置为利用双目图像中第一图像的第一深度图对第二图像执行对齐处理,得到所述第一图像的校准图,以及利用所述第二图像的第一深度图对所述第一图像执行对齐处理,得到所述第二图像的校准图;根据双目图像中各图像与对应的校准图之间的差异,分别得到所述第一图像和第二图像的掩模图。
在一些可能的实施方式中,所述融合特征模块,还配置为按照第一预设方式,基于所述第一图像的校准图,以及所述第一图像的掩模图得到所述第一图像的中间融合特征图;以及按照第二预设方式,基于所述第二图像的校准图,以及所述第二图像的掩模图得到所述第二图像的中间融合特征图。
在一些可能的实施方式中,所述第一预设方式的表达式为:
Figure PCTCN2019113749-appb-000005
其中,
Figure PCTCN2019113749-appb-000006
表示为第一图像的中间融合特征,⊙表示对应元素相乘,W L(I R)表示利用第一图像的第一深度图执行第二图像的对其处理 后的结果,M L表示第一图像的掩模图;
所述第二预设方式的表达式为:
Figure PCTCN2019113749-appb-000007
其中,
Figure PCTCN2019113749-appb-000008
表示为第二图像的中间融合特征,⊙表示对应元素相乘,W R(F L)表示利用第二图像的第一深度图执行第一图像的对齐处理后的结果,M R表示第二图像的掩模图。
在一些可能的实施方式中,所述优化模块还用于分别对所述双目图像的融合特征图执行卷积处理,得到所述去模糊处理后的双目图像。
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。电子设备可以被提供为终端、服务器或其它形态的设备。
本申请实施例公开了一种计算机程序产品,所述计算机程序产品包括计算机程序指令,其中,所述计算机程序指令被处理器执行时实前述任意方法。
图11示出根据本公开实施例的一种电子设备800的框图。例如,电设备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11,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组件实现,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12示出根据本公开实施例的一种电子设备1900的框图。例如,电子设备1900可以被提供为一服务器。参照图12,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,以及常规的过程序编程语言—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 (29)

  1. 一种图像处理方法,包括:
    获取双目图像,其中,所述双目图像包括针对同一对象在同一场景下拍摄的第一图像和第二图像;
    获得所述双目图像的第一特征图、所述双目图像的第一深度图,以及融合所述双目图像的图像特征和深度特征的第二特征图;
    对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图;
    对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像。
  2. 根据权利要求1所述的方法,其中,所述获得所述双目图像的第一特征图,包括:
    对所述第一图像和第二图像分别执行第一卷积处理,得到所述第一图像和第二图像分别对应的第一中间特征图;
    对所述第一图像和第二图像的所述第一中间特征图分别执行第二卷积处理,得到所述第一图像和第二图像分别对应的多尺度的第二中间特征图;
    对所述第一图像和第二图像的各尺度的第二中间特征图分别执行残差处理,得到所述第一图像和第二图像分别对应的第一特征图。
  3. 根据权利要求2所述的方法,其中,所述对所述双目图像的第一图像和第二图像分别执行第一卷积处理,得到所述第一图像和第二图像分别对应的第一中间特征图,包括:
    利用第一预设卷积核以及第一卷积步长分别对所述第一图像和第二图像分别执行卷积处理,得到所述第一图像和第二图像分别对应的第一中间特征图。
  4. 根据权利要求2或3所述的方法,其中,所述对所述第一图像和第二图像的所述第一中间特征图分别执行第二卷积处理,得到所述第一图像和第二图像分别对应的多尺度的第二中间特征图,包括:
    分别按照预设的多个不同的第一空洞率,对所述第一图像和第二图像的所述第一中间特征图执行卷积处理,得到与该多个第一空洞率分别对应的第二中间特征图。
  5. 根据权利要求2至4中任意一项所述的方法,其中,所述对所述第一图像和第二图像的各尺度的第二中间特征图分别执行残差处理,得到所述第一图像和第二图像分别对应的第一特征图,包括:
    分别连接所述第一图像的多个尺度的第二中间特征图得到第一连接特征图,以及分别连接第二图像的多个尺度的第二中间特征图得到第二连接特征图;
    分别对所述第一连接特征图和第二连接特征图执行卷积处理;
    对所述第一图像的第一中间特征图和卷积处理后的第一连接特征图执行相加处理,得到第一图像的第一特征图,以及对所述第二图像的第一中间特征图和卷积处理后的第二连接特征图执行相加处理,得到所述第二图像的第一特征图。
  6. 根据权利要求1至5中任意一项所述的方法,其中,获得所述双目图像的第一深度图,以及融合所述双目图像的图像特征和深度特征的第二特征图,包括;
    将所述第一图像和第二图像进行组合,形成组合视图;
    对所述组合视图执行至少一层第三卷积处理得到第一中间深度特征图;
    对所述第一中间深度特征图执行第四卷积处理,得到多个尺度的第二中间深度特征图;
    对所述第二中间深度特征与所述第一中间深度图执行残差处理,分别得到所述第一图像和第二图像的第一深度图,以及根据任意一层第三卷积处理获得所述第二特征图。
  7. 根据权利要求6所述的方法,其中,所述对所述组合视图执行至少一层第三卷积 处理得到第一中间深度特征图,包括:
    利用第二预设卷积核以及第二卷积步长对所述组合视图执行至少一次卷积处理,得到所述第一中间深度特征图。
  8. 根据权利要求6或7所述的方法,其中,所述对所述第一中间深度特征图执行第四卷积处理,得到多个尺度的第二中间深度特征图,包括:
    分别按照预设的多个不同的第二空洞率,对所述第一中间深度特征图执行卷积处理,得到与该多个第二空洞率分别对应的第二中间深度特征图。
  9. 根据权利要求1至8中任意一项所述的方法,其中,所述对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图,包括:
    根据所述双目图像中第一图像的第一深度图对第二图像执行校准处理,获得所述第一图像掩模图,以及根据所述双目图像中第二图像的第一深度图对第一图像执行校准处理,获得所述第二图像的掩模图;
    基于所述双目图像中各图像对应的所述校准图和掩模图,分别获得所述双目图像中各图像的中间融合特征;
    根据所述双目图像中各图像的第一深度图和第二特征图,获得所述双目图像各图像的深度特征融合图;
    根据所述双目图像中各图像的第一图像的第一特征图、第一图像的中间融合特征图以及第一图像的深度特征融合图的连接结果,对应的得到各图像的所述融合特征图。
  10. 根据权利要求9所述的方法,其中,所述根据所述双目图像中第一图像的第一深度图对第二图像执行校准处理,获得所述第一图像掩模图,以及根据所述双目图像中第二图像的第一深度图对第一图像执行校准处理,获得所述第二图像的掩模图,包括:
    利用双目图像中第一图像的第一深度图对第二图像执行对齐处理,得到所述第一图像的校准图,以及利用所述第二图像的第一深度图对所述第一图像执行对齐处理,得到所述第二图像的校准图;
    根据双目图像中各图像与对应的校准图之间的差异,分别得到所述第一图像和第二图像的掩模图。
  11. 根据权利要求9所述的方法,其中,基于所述双目图像中各图像对应的所述校准图和掩模图,分别获得所述双目图像中各图像的中间融合特征,包括:
    按照第一预设方式,基于所述第一图像的校准图,以及所述第一图像的掩模图得到所述第一图像的中间融合特征图;以及
    按照第二预设方式,基于所述第二图像的校准图,以及所述第二图像的掩模图得到所述第二图像的中间融合特征图。
  12. 根据权利要求11所述的方法,其中,所述第一预设方式的表达式为:
    Figure PCTCN2019113749-appb-100001
    其中,
    Figure PCTCN2019113749-appb-100002
    表示为第一图像的中间融合特征,⊙表示对应元素相乘,W L(I R)表示利用第一图像的第一深度图执行第二图像的对其处理后的结果,M L表示第一图像的掩模图;
    所述第二预设方式的表达式为:
    Figure PCTCN2019113749-appb-100003
    其中,
    Figure PCTCN2019113749-appb-100004
    表示为第二图像的中间融合特征,⊙表示对应元素相乘,W R(F L)表示利用第二图像的第一深度图执行第一图像的对齐处理后的结果,M R表示第二图像的掩模图。
  13. 根据权利要求1-12中任意一项所述的方法,其中,所述对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像,包括:
    分别对所述双目图像的融合特征图执行卷积处理,得到所述去模糊后处理的双目图像。
  14. 一种图像处理装置,其中,包括:
    获取模块,配置为获取双目图像,其中,所述双目图像包括针对同一对象在同一场景下拍摄的第一图像和第二图像;
    特征提取模块,配置为获得所述双目图像的第一特征图、所述双目图像的第一深度图,以及融合所述双目图像的图像特征和深度特征的第二特征图;
    特征融合模块,配置为对所述双目图像、所述双目图像的第一特征图、第一深度图以及所述第二特征图进行特征融合处理,得到所述双目图像的融合特征图;
    优化模块,配置为对所述双目图像的融合特征图执行优化处理,得到去模糊处理后的双目图像。
  15. 根据权利要求14所述的装置,其中,所述特征提取模块包括图像特征提取模块,配置为对所述第一图像和第二图像分别执行第一卷积处理,得到所述第一图像和第二图像分别对应的第一中间特征图;
    对所述第一图像和第二图像的所述第一中间特征图分别执行第二卷积处理,得到所述第一图像和第二图像分别对应的多尺度的第二中间特征图;以及
    对所述第一图像和第二图像的各尺度的第二中间特征图分别执行残差处理,得到所述第一图像和第二图像分别对应的第一特征图。
  16. 根据权利要求15所述的装置,其中,所述图像特征提取模块,还配置为利用第一预设卷积核以及第一卷积步长分别对所述第一图像和第二图像分别执行卷积处理,得到所述第一图像和第二图像分别对应的第一中间特征图。
  17. 根据权利要求15或16所述的装置,其中,所述图像特征提取模块还用于分别按照预设的多个不同的第一空洞率,对所述第一图像和第二图像的所述第一中间特征图执行卷积处理,得到与该多个第一空洞率分别对应的第二中间特征图。
  18. 根据权利要求15至17中任意一项所述的装置,其中,所述图像特征提取模块,还配置为分别连接所述第一图像的多个尺度的第二中间特征图得到第一连接特征图,以及分别连接第二图像的多个尺度的第二中间特征图得到第二连接特征图;
    分别对所述第一连接特征图和第二连接特征图执行卷积处理;以及
    对所述第一图像的第一中间特征图和卷积处理后的第一连接特征图执行相加处理,得到第一图像的第一特征图,以及对所述第二图像的第一中间特征图和卷积处理后的第二连接特征图执行相加处理,得到所述第二图像的第一特征图。
  19. 根据权利要求14至18中任意一项所述的装置,其中,所述特征提取模块还包括深度特征提取模块,配置为将所述第一图像和第二图像进行组合,形成组合视图;
    对所述组合视图执行至少一层第三卷积处理得到第一中间深度特征图;
    对所述第一中间深度特征图执行第四卷积处理,得到多个尺度的第二中间深度特征图;以及
    对所述第二中间深度特征与所述第一中间深度图执行残差处理,分别得到所述第一图像和第二图像的第一深度图,以及根据任意一层第三卷积处理获得所述第二特征图。
  20. 根据权利要求19所述的装置,其中,所述深度特征提取模块,还配置为利用第二预设卷积核以及第二卷积步长对所述组合视图执行至少一次卷积处理,得到所述第一中间深度特征图。
  21. 根据权利要求19或20所述的装置,其中,所述深度特征提取模块还配置为分别按照预设的多个不同的第二空洞率,对所述第一中间深度特征图执行卷积处理,得到与该多个第二空洞率分别对应的第二中间深度特征图。
  22. 根据权利要求14-21中任意一项所述的装置,其中,所述特征融合模块,还配置为根据所述双目图像中第一图像的第一深度图对第二图像执行校准处理,获得所述第一图像掩模图,以及根据所述双目图像中第二图像的第一深度图对第一图像执行校准处理,获得所述第二图像的掩模图;
    基于所述双目图像中各图像对应的所述校准图和掩模图,分别获得所述双目图像中各图像的中间融合特征;
    根据所述双目图像中各图像的第一深度图和第二特征图,获得所述双目图像各图像的深度特征融合图;以及
    根据所述双目图像中各图像的第一图像的第一特征图、第一图像的中间融合特征图以及第一图像的深度特征融合图的连接结果,对应的得到各图像的所述融合特征图。
  23. 根据权利要求22所述的装置,其中,所述特征融合模块,还配置为利用双目图像中第一图像的第一深度图对第二图像执行对齐处理,得到所述第一图像的校准图,以及利用所述第二图像的第一深度图对所述第一图像执行对齐处理,得到所述第二图像的校准图;
    根据双目图像中各图像与对应的校准图之间的差异,分别得到所述第一图像和第二图像的掩模图。
  24. 根据权利要求22所述的装置,其中,所述融合特征模块,还配置为按照第一预设方式,基于所述第一图像的校准图,以及所述第一图像的掩模图得到所述第一图像的中间融合特征图;以及
    按照第二预设方式,基于所述第二图像的校准图,以及所述第二图像的掩模图得到所述第二图像的中间融合特征图。
  25. 根据权利要求24所述的装置,其中,所述第一预设方式的表达式为:
    Figure PCTCN2019113749-appb-100005
    其中,
    Figure PCTCN2019113749-appb-100006
    表示为第一图像的中间融合特征,⊙表示对应元素相乘,W L(I R)表示利用第一图像的第一深度图执行第二图像的对其处理后的结果,M L表示第一图像的掩模图;
    所述第二预设方式的表达式为:
    Figure PCTCN2019113749-appb-100007
    其中,
    Figure PCTCN2019113749-appb-100008
    表示为第二图像的中间融合特征,⊙表示对应元素相乘,W R(F L)表示利用第二图像的第一深度图执行第一图像的对齐处理后的结果,M R表示第二图像的掩模图。
  26. 根据权利要求14-23中任意一项所述的装置,其中,所述优化模块还用于分别对所述双目图像的融合特征图执行卷积处理,得到所述去模糊处理后的双目图像。
  27. 一种电子设备,其中,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13中任意一项所述的方法。
  28. 一种计算机可读存储介质,其上存储有计算机程序指令,其中,所述计算机程序指令被处理器执行时实现权利要求1至13中任意一项所述的方法。
  29. 一种计算机程序产品,所述计算机程序产品包括计算机程序指令,其中,所述计算机程序指令被处理器执行时实现权利要求1至13中任意一项所述的方法。
PCT/CN2019/113749 2019-01-22 2019-10-28 图像处理方法及装置、电子设备和存储介质 WO2020151281A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
JP2020573341A JP7033674B2 (ja) 2019-01-22 2019-10-28 画像処理方法及び装置、電子機器並びに記憶媒体
SG11202106271XA SG11202106271XA (en) 2019-01-22 2019-10-28 Image processing method and device, electronic equipment and storage medium
KR1020217002881A KR20210028218A (ko) 2019-01-22 2019-10-28 이미지 처리 방법과 장치, 전자 기기 및 저장 매체
US17/345,042 US20210319538A1 (en) 2019-01-22 2021-06-11 Image processing method and device, electronic equipment and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910060238.6 2019-01-22
CN201910060238.6A CN109829863B (zh) 2019-01-22 2019-01-22 图像处理方法及装置、电子设备和存储介质

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/345,042 Continuation US20210319538A1 (en) 2019-01-22 2021-06-11 Image processing method and device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
WO2020151281A1 WO2020151281A1 (zh) 2020-07-30
WO2020151281A9 true WO2020151281A9 (zh) 2020-09-10

Family

ID=66861908

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/113749 WO2020151281A1 (zh) 2019-01-22 2019-10-28 图像处理方法及装置、电子设备和存储介质

Country Status (7)

Country Link
US (1) US20210319538A1 (zh)
JP (1) JP7033674B2 (zh)
KR (1) KR20210028218A (zh)
CN (1) CN109829863B (zh)
SG (1) SG11202106271XA (zh)
TW (1) TWI706379B (zh)
WO (1) WO2020151281A1 (zh)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109829863B (zh) * 2019-01-22 2021-06-25 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
CN109977847B (zh) * 2019-03-22 2021-07-16 北京市商汤科技开发有限公司 图像生成方法及装置、电子设备和存储介质
CN110060215B (zh) * 2019-04-16 2021-09-10 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
WO2020238120A1 (en) * 2019-05-30 2020-12-03 Guangdong Oppo Mobile Telecommunications Corp., Ltd. System and method for single-modal or multi-modal style transfer and system for random stylization using the same
CN110543849B (zh) * 2019-08-30 2022-10-04 北京市商汤科技开发有限公司 检测器的配置方法及装置、电子设备和存储介质
CN111986075B (zh) * 2020-08-12 2022-08-09 兰州交通大学 一种目标边缘清晰化的风格迁移方法
CN112101302B (zh) * 2020-11-05 2021-04-27 杭州追猎科技有限公司 非法张贴物检测方法、系统和电子设备
US11669986B2 (en) * 2021-04-16 2023-06-06 Adobe Inc. Generating enhanced three-dimensional object reconstruction models from sparse set of object images
KR102554665B1 (ko) * 2021-12-20 2023-07-12 포항공과대학교 산학협력단 역커널 기반 디포커스 블러 제거 방법 및 장치
KR102664276B1 (ko) * 2022-12-02 2024-05-10 고려대학교 산학협력단 수직-수평 분할을 통한 단일 영상 잔상 제거 방법 및 이를 위한 네트워크 시스템
CN116862800B (zh) * 2023-07-11 2024-01-30 哈尔滨工业大学 一种大视场单透镜空变模糊图像复原方法和装置

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8380630B2 (en) * 2000-07-06 2013-02-19 David Paul Felsher Information record infrastructure, system and method
US20140198977A1 (en) * 2012-03-21 2014-07-17 Texas Instruments Incorporated Enhancement of Stereo Depth Maps
CN105516579B (zh) * 2014-09-25 2019-02-05 联想(北京)有限公司 一种图像处理方法、装置和电子设备
JP6929047B2 (ja) * 2016-11-24 2021-09-01 キヤノン株式会社 画像処理装置、情報処理方法及びプログラム
US10409614B2 (en) * 2017-04-24 2019-09-10 Intel Corporation Instructions having support for floating point and integer data types in the same register
US10489877B2 (en) * 2017-04-24 2019-11-26 Intel Corporation Compute optimization mechanism
CN107392868A (zh) * 2017-07-21 2017-11-24 深圳大学 基于全卷积神经网络的压缩双目图像质量增强方法及装置
CN108269280A (zh) * 2018-01-05 2018-07-10 厦门美图之家科技有限公司 一种深度图像的处理方法及移动终端
CN108596040A (zh) * 2018-03-29 2018-09-28 中山大学 一种基于双目视觉的串联通道融合行人检测方法
CN108765333B (zh) * 2018-05-24 2021-08-10 华南理工大学 一种基于深度卷积神经网络的深度图完善方法
CN108846820A (zh) * 2018-07-10 2018-11-20 深圳市唯特视科技有限公司 一种基于尺度递归网络的深度图像去模糊方法
CN109118430B (zh) * 2018-08-24 2023-05-09 深圳市商汤科技有限公司 超分辨率图像重建方法及装置、电子设备及存储介质
CN109829863B (zh) * 2019-01-22 2021-06-25 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
CN110766623A (zh) * 2019-10-12 2020-02-07 北京工业大学 一种基于深度学习的立体图像修复方法

Also Published As

Publication number Publication date
KR20210028218A (ko) 2021-03-11
US20210319538A1 (en) 2021-10-14
SG11202106271XA (en) 2021-07-29
JP2021530056A (ja) 2021-11-04
CN109829863A (zh) 2019-05-31
CN109829863B (zh) 2021-06-25
JP7033674B2 (ja) 2022-03-10
TWI706379B (zh) 2020-10-01
TW202029125A (zh) 2020-08-01
WO2020151281A1 (zh) 2020-07-30

Similar Documents

Publication Publication Date Title
TWI706379B (zh) 圖像處理方法及裝置、電子設備和儲存介質
TWI777162B (zh) 圖像處理方法及裝置、電子設備和電腦可讀儲存媒體
JP7262659B2 (ja) 目標対象物マッチング方法及び装置、電子機器並びに記憶媒体
WO2021196401A1 (zh) 图像重建方法及装置、电子设备和存储介质
EP3326360B1 (en) Image capturing apparatus and method of operating the same
JP7072119B2 (ja) 画像処理方法および装置、電子機器ならびに記憶媒体
CN109788189B (zh) 将相机与陀螺仪融合在一起的五维视频稳定化装置及方法
KR102463101B1 (ko) 이미지 처리 방법 및 장치, 전자 기기 및 저장 매체
EP3057304B1 (en) Method and apparatus for generating image filter
WO2017124899A1 (zh) 一种信息处理方法及装置、电子设备
WO2021189704A1 (zh) 图像处理方法及装置、电子设备和存储介质
WO2018026586A1 (en) Combining images aligned to reference frame
TWI778313B (zh) 圖像處理方法、電子設備和儲存介質
WO2021031790A1 (zh) 一种信息处理方法、装置、电子设备、存储介质和程序
WO2022193507A1 (zh) 图像处理方法及装置、设备、存储介质、程序和程序产品
CN115134532A (zh) 图像处理方法、装置、存储介质及电子设备
CN109543544B (zh) 跨光谱图像匹配方法及装置、电子设备和存储介质
Anthes Smarter photography
CN113744126A (zh) 图像处理方法及装置、计算机可读介质和电子设备
CN116416141A (zh) 图像处理方法及装置、设备、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: 19912005

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020573341

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20217002881

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19912005

Country of ref document: EP

Kind code of ref document: A1

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 09.11.2021)

122 Ep: pct application non-entry in european phase

Ref document number: 19912005

Country of ref document: EP

Kind code of ref document: A1