WO2022022350A1 - 图像处理方法及装置、电子设备、存储介质和计算机程序产品 - Google Patents

图像处理方法及装置、电子设备、存储介质和计算机程序产品 Download PDF

Info

Publication number
WO2022022350A1
WO2022022350A1 PCT/CN2021/107512 CN2021107512W WO2022022350A1 WO 2022022350 A1 WO2022022350 A1 WO 2022022350A1 CN 2021107512 W CN2021107512 W CN 2021107512W WO 2022022350 A1 WO2022022350 A1 WO 2022022350A1
Authority
WO
WIPO (PCT)
Prior art keywords
target object
image
dimensional image
target
path
Prior art date
Application number
PCT/CN2021/107512
Other languages
English (en)
French (fr)
Inventor
刘畅
赵亮
Original Assignee
上海商汤智能科技有限公司
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 上海商汤智能科技有限公司 filed Critical 上海商汤智能科技有限公司
Publication of WO2022022350A1 publication Critical patent/WO2022022350A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds

Definitions

  • the present disclosure relates to the field of computer technology, and in particular, to an image processing method, apparatus, electronic device, storage medium, and computer program product.
  • the target recognition method based on neural network can use the self-learning characteristics of neural network to extract image features and identify the target object in the image.
  • the recognition accuracy of the target object in the three-dimensional image needs to be improved.
  • the present disclosure proposes an image processing technical solution.
  • an image processing method comprising:
  • the target object is identified in the three-dimensional image, and the position of the target object in the three-dimensional image is determined.
  • identifying the key features and extension directions of the target object in the three-dimensional image includes:
  • the extension direction of the reference object of the target object is identified, and the extension direction of the reference object of the target object is taken as the extension direction of the target object, and the size of the reference object is larger than the size of the target object.
  • the extension direction of the target object may be determined by identifying the extension direction of the reference object with a larger size.
  • the identifying the target object in the three-dimensional image according to the key feature and the extending direction includes:
  • the target area of the 3D image can be resampled to obtain a resampled image, while other areas in the 3D image that are not related to target object recognition may not be resampled, which can improve the efficiency of target object recognition.
  • the target area containing the target object is resampled to obtain a resampled image, including:
  • the target area of the rotated three-dimensional image is resampled to obtain a resampled image.
  • the three-dimensional image can be rotated so that the extending direction of the target object in the rotated three-dimensional image is parallel to the sampling plane.
  • the plane where the target object is located in the rotated three-dimensional image coincides with the sampling plane
  • resampling the target area of the rotated three-dimensional image according to the sampling plane to obtain a resampled image including:
  • the plane where the target object is located in the rotated three-dimensional image coincides with the sampling plane, so that complete sampling of the target object can be achieved in the case of resampling.
  • the resampled image is a three-dimensional image
  • the method further includes:
  • the performing target object recognition on the resampled image to determine the position of the target object in the three-dimensional image includes:
  • the coordinates of the target object in the three-dimensional image are determined.
  • the recognition result of the 3D image after rotation can be mapped to the 3D image before rotation, so that the position of the target object in the original 3D image can be determined, which can facilitate the user to view the position of the target object in the original 3D image. Improved user experience.
  • the target object is in the shape of a line
  • performing target object recognition on the resampled image to determine the position of the target object in the three-dimensional image includes:
  • the position of the linear feature and the target path is taken as the position of the target object.
  • the target paths that satisfy the preset conditions in the paths connecting the linear features are also regarded as part of the target objects, so as to solve the problem that the identified target objects are discontinuous. problem, and improve the accuracy of target object recognition.
  • the target object includes a dental neural canal
  • the reference object of the target object includes at least two teeth adjacent to the dental neural canal
  • the three-dimensional images include cone beam computed tomography images CBCT.
  • the position of the dental nerve canal in the CBCT image can be determined accurately and quickly.
  • the determining of a target path in the path that satisfies a preset condition includes:
  • the cumulative energy function is related to at least one of the following factors: the sum of the grayscale values of the pixels on the path, the path length, and the smoothness of the path;
  • the value of the cumulative energy function is positively correlated with the sum of the grayscale values of the pixels on the path, positively correlated with the path length, and negatively correlated with the smoothness.
  • the path with the lowest value of the accumulated energy function among the paths connecting the linear features is also taken as a part of the target object, and the identified target object is solved.
  • the discontinuous problem improves the accuracy of target object recognition.
  • the target area includes a to-be-implanted area in the oral cavity and an area where teeth adjacent to the to-be-implanted area are located.
  • the determined target area usually includes the teeth adjacent to the area to be implanted (ie, the missing tooth position), and the dental neural canal connected with the tooth, including the dental neural canal at the gum corresponding to the missing tooth position.
  • an image processing apparatus including:
  • an identification module configured to identify key features and extension directions of the target object in the three-dimensional image
  • the position determination module is configured to identify the target object in the three-dimensional image according to the key feature and the extension direction, and determine the position of the target object in the three-dimensional image.
  • an electronic device comprising: a processor; a memory configured to store instructions executable by the processor; wherein the processor is configured to invoke the instructions stored in the memory to execute the above method.
  • a computer-readable storage medium having computer program instructions stored thereon, the computer program instructions implementing the above method when executed by a processor.
  • a computer program product comprising one or more instructions adapted to be loaded and executed by a processor to implement the above method.
  • the target object in the three-dimensional image can be identified according to the extension direction in the three-dimensional image, Since the target object exists in the extension direction, other regions other than the extension direction may not be recognized, which can improve the efficiency of target object recognition.
  • FIG. 1A shows a flowchart of an image processing method according to an embodiment of the present disclosure
  • FIG. 1B shows a flowchart of an algorithm for automatic segmentation of the mandibular dental neural canal according to an embodiment of the present disclosure
  • FIG. 2 shows a block diagram of an image processing apparatus according to an embodiment of the present disclosure
  • FIG. 3 shows a block diagram of an electronic device according to an embodiment of the present disclosure
  • FIG. 4 shows a block diagram of an electronic device according to an embodiment of the present disclosure.
  • FIG. 1A shows a flowchart of an image processing method according to an embodiment of the present disclosure. As shown in FIG. 1A , the image processing method includes:
  • step S11 key features and extension directions of the target object in the three-dimensional image are identified.
  • Three-dimensional images can also be called stereo images.
  • Three-dimensional images can be collected from real objects through stereo image acquisition equipment, such as three-dimensional images of human organs collected in medical diagnosis, three-dimensional images of products, etc.
  • Three-dimensional images drawn by drawing techniques. The present disclosure does not limit the form and source of the three-dimensional image.
  • the key features of the target object are used to characterize the features of the key parts of the target object and are part of the target object. Compared with the features of other parts of the target object, the key features are easier to identify. That is, the key feature of the target object may be a feature of a part of the target object that is easily recognized, for example, a feature of a part of the target object whose size is larger than a set size threshold.
  • the initial position of the dental neural canal is often thicker, and the distal position of the dental neural canal is often thinner. Therefore, the characteristics of the initial position of the dental neural canal can be used. as the key feature of the target object.
  • the identification of key features may be performed through a trained network, and the identification method is not limited in this disclosure. Since the key features of the target object are larger in size than other parts, they are easy to identify. In the process of identifying the target object, image processing algorithms with lower complexity can be used for identification to improve the identification efficiency.
  • the extension direction of the target object may be the direction in which the target object extends in the 3D image with the key feature as the starting point. Since the target object may only extend in one direction, or may extend in multiple directions, the direction may be It can be unidirectional or multidirectional, or it can be a certain plane extending around the key feature as the center point.
  • step S12 the target object is identified in the three-dimensional image according to the key feature and the extension direction, and the position of the target object in the three-dimensional image is determined.
  • the target object in the three-dimensional image can be identified along the extension direction of the key feature in the three-dimensional image. Other areas of the target object can not be recognized, which can greatly improve the efficiency of target object recognition.
  • the identifying the key features and extension directions of the target object in the three-dimensional image includes: identifying the reference object of the target object.
  • the extension direction of the reference object of the target object is taken as the extension direction of the target object, and the size of the reference object is larger than the size of the target object.
  • the extension direction of the target object can be determined by identifying the extension direction of the reference object with a larger size.
  • the reference object is a pre-selected object that extends in the same direction as the target object.
  • the reference objects of the target object may be at least two teeth adjacent to the dental neural canal, since the extension direction of the at least two teeth adjacent to the dental neural canal is different from the dental nerve The extension direction of the canal is the same, therefore, at least two teeth adjacent to the dental nerve canal can be used as reference objects of the target object.
  • the size of the teeth is relatively large, the extending directions of the two teeth can be easily determined, thereby improving the efficiency of identifying the extending directions.
  • the present disclosure also provides an implementation method for identifying the extension direction of the target object.
  • the target object in the three-dimensional image can be roughly identified, so that although the full picture of the target object cannot be obtained, However, multiple target features of the target object can be obtained, and then the extension direction formed by connecting the multiple target features is determined as the extension direction of the target object.
  • the three-dimensional image in the rough identification process, may be downsampled by a sampling rate lower than the set threshold, and the target object identification may be performed on the image obtained after the downsampling. Since the sampling rate is low, therefore, Although the whole picture of the target object cannot be obtained, the extension direction of the target object can be determined according to the obtained multiple target features. Through this implementation, the general position of the target object can be quickly determined.
  • the identifying the target object in the three-dimensional image according to the key features and the extending direction includes: according to the key features and all the three-dimensional images in the three-dimensional image.
  • the extension direction is to resample the target area including the target object to obtain a resampled image, and the resampled sampling plane is parallel to the extension direction of the target object; the resampled image is subjected to target object recognition to determine the position of the target object in the three-dimensional image.
  • the extending direction starting from the key feature is the target area containing the target object. Therefore, in the case of identifying the target object, you can Identify the target area.
  • the target area of the 3D image may be resampled to obtain a resampled image, while other areas in the 3D image that are not related to target object recognition may not be resampled, which can improve the efficiency of target object recognition.
  • the sampling plane can be kept parallel to the extension direction of the target object.
  • the sampling plane can be the plane where a 2D image is collected at a time during the sampling of the image.
  • the sampling plane can be the yz plane of the 3D image, while the target object is on the yz plane.
  • the extension direction of is also in the yz plane, the target object can be collected into a two-dimensional image.
  • image processing technologies such as image recognition technology, image semantic segmentation technology, and target detection technology can be used to determine the position of the target object in the resampling image. Therefore, the spatial mapping relationship between the resampling image and the three-dimensional graphics can be established, and then the position of the target object in the three-dimensional image can be determined according to the recognition result of the resampling image.
  • a resampled image is obtained by determining a target area containing the target object to be identified in the three-dimensional image, and then keeping the sampling plane parallel to the extending direction of the target object during the process of resampling the target area; Target object recognition is performed on the resampling image, and the position of the target object in the three-dimensional image is determined.
  • the sampling plane parallel to the extension direction of the target object the target object can be captured in a whole resampled image as much as possible, so that the complete target object in the resampled image can be segmented.
  • the proportion of target object pixels in the obtained resampling image is increased, which reduces the influence of the data imbalance problem on the recognition accuracy, and improves the accuracy of the recognized target object.
  • re-sampling the target area to obtain a re-sampled image includes: rotating the three-dimensional image, and the extending direction of the target object in the rotated three-dimensional image is parallel to the sampling plane ; According to the sampling plane, re-sampling the target area of the rotated three-dimensional image to obtain a re-sampled image.
  • the 3D image After determining the extension direction of the target object, while keeping the sampling plane of the sampler unchanged, the 3D image can be rotated so that the extension direction of the target object in the rotated 3D image is parallel to the sampling plane.
  • the target object can often be completely contained in the resampling image.
  • the target object can be completely segmented, and the segmentation accuracy of the target object can be improved.
  • the resampling image may be a two-dimensional image or a three-dimensional image, that is, in the process of resampling the target area, a two-dimensional image or a three-dimensional image may be collected.
  • the plane where the target object is located in the rotated three-dimensional image coincides with the sampling plane, so that both can be achieved in the case of resampling A full sample of the target object.
  • the plane on which the target object is located can be determined first, and after the plane on which the target object is located is determined, the sampling plane can be used as the sampling plane for the graphics. Do resampling.
  • the target object can often be completely contained in the resampling image.
  • performing target recognition on the resampling image can completely recognize the target object, improve the accuracy of the recognized target object, and reduce the poor target object recognition results when the target object is not parallel to the sampling plane.
  • the resampled image is a 3D image
  • resample the target area of the rotated 3D image to obtain a resampled image including: Resampling the target area of the rotated 3D image along a direction perpendicular to the sampling plane to obtain a 3D resampling image including the target object.
  • the plane formed by the directions of the two longer dimensions is the extension direction of the three-dimensional target object.
  • the 3D image After determining the plane where the target object is located, while keeping the sampling plane of the sampler unchanged, the 3D image can be rotated so that the extension direction of the target object in the rotated 3D image is parallel to the sampling plane. After rotation, the proportion of target object pixels in the obtained resampling image is increased, which reduces the influence of the data imbalance problem on the recognition accuracy, and improves the accuracy of the recognized target object.
  • the method further includes: determining a mapping relationship between the resampled image and the spatial coordinates of the three-dimensional image according to the rotation angle of the three-dimensional image; the Performing target object recognition on the resampled image, and determining the position of the target object in the three-dimensional image, includes: performing semantic segmentation on the target object in the resampled image to obtain the target object in the resampled image. Position information in the sampled image; according to the position information of the target object in the resampled image and the mapping relationship, determine the coordinates of the target object in the three-dimensional image.
  • Semantic segmentation is a classification at the pixel level of the image. Pixels belonging to the same category in the image will be classified into one category. By semantically segmenting the target object in the resampling image, the target object in the resampling image will be obtained. A class of pixels, the position information of the pixel of the target object is the position information of the target object in the resampling image.
  • the process of performing semantic segmentation can be implemented by a trained network, and the network can be obtained by training with labeled samples, which will not be described in detail in this disclosure.
  • a certain point in the three-dimensional space can be used as the rotation center, the three-dimensional image can be rotated, and the rotation angle of the three-dimensional image relative to the rotation center can be determined. Then, according to the rotation angle of the three-dimensional image, the mapping relationship between the sampling image and the spatial coordinates of the three-dimensional image is determined.
  • the mapping relationship may be a pixel point in the three-dimensional image, and the corresponding relationship between coordinates before and after rotation.
  • the recognition result of the three-dimensional image after rotation can be mapped to the three-dimensional image before rotation, that is, The position of the target object in the original three-dimensional image can be determined, it is convenient for the user to view the position of the target object in the original three-dimensional image, and the user experience is improved.
  • the target object is in the shape of a line
  • performing target object recognition on the resampled image to determine the position of the target object in the three-dimensional image includes: When the linear features of the target object are discontinuous, determine a path connecting the linear features in the three-dimensional image; determine a target path in the path that satisfies a preset condition; combine the linear features with the The position of the target path is taken as the position of the target object.
  • the part of the target object exists on the image but is not recognized, or the part of the target object is not clear enough in the image to be recognized or even invisible.
  • the identified linear features of the target object are discontinuous, paths connecting the identified linear features can be determined, and there are unrecognized linear features in these paths. Therefore, a target path satisfying a preset condition can be selected from these paths as a linear feature not recognized in the target recognition process.
  • the target path that satisfies the preset conditions in the paths connecting the linear features is also regarded as a part of the target object, so as to solve the problem of identifying the The target object is discontinuous, which improves the accuracy of target object recognition.
  • the target object includes a dental neural canal; and the three-dimensional image includes a cone beam computed tomography (CBCT).
  • CBCT cone beam computed tomography
  • the target path of the preset condition may be one of the paths between two adjacent linear features.
  • the path with the lowest value of the cumulative energy function then, determining the target path in the path that satisfies the preset condition may include: taking the path with the lowest value of the cumulative energy function among the paths between two adjacent linear features as the phase The target path between two adjacent linear features.
  • the cumulative energy function is related to at least one of the following factors: the sum of the grayscale values of the pixels on the path; the path length; the smoothness of the path;
  • the value of the cumulative energy function is positively correlated with the sum of the grayscale values of the pixels on the path, positively correlated with the path length, and negatively correlated with the smoothness.
  • the path of the dental neural canal is closer to a straight line. Therefore, the shorter the path, the lower the value of the cumulative energy function, and the greater the probability that the path is the dental neural canal.
  • the smoothness of the dental neural canal is often higher. Therefore, the higher the smoothness, the lower the value of the cumulative energy function, and the greater the probability that the path is the dental neural canal.
  • the path with the lowest value of the accumulated energy function among the paths connecting the linear features is also used as a part of the target object to solve the problem.
  • the problem that the identified target objects are discontinuous improves the accuracy of target object recognition.
  • the target area containing the target object to be identified includes the area to be implanted in the oral cavity and the area where the teeth adjacent to the area to be implanted are located.
  • the target area thus determined usually includes the teeth adjacent to the area to be implanted (ie, the missing tooth position), and the dental neural canal connected with the tooth, including the dental neural canal at the gum corresponding to the missing tooth position.
  • Dental implants are the most common way to restore missing teeth. In implant placement surgery, the placement of the implant directly affects the success of the surgery. In the surgical planning, the implant position of the dental implant should avoid the dental neural canal located in the gums, so as to avoid damage and compression to the dental neural canal.
  • the position of the dental neural canal in the CBCT image can be determined accurately and quickly.
  • the area to be implanted and the area of the teeth adjacent to the area to be implanted in the CBCT oral image are determined. ; Then resample the area to obtain a resampled image, in which the sampling plane is parallel to the extension direction of the dental neural canal by rotating the CBCT image, so that the dental neural canal can be collected as completely as possible; then the sampling The image is used to identify the dental neural canal.
  • the shortest path between the identified adjacent dental neural canal segments can be used to complete it, and it can be determined that the dental neural canal is in the resampling image. Then, according to the mapping relationship, the position of the dental neural canal in the original CBCT image was determined.
  • the CBCT image it is often the image of the entire oral cavity and its surrounding areas, while the number of pixels occupied by the dental neural canal is small, and the proportion of pixels is low. In the process, there is often a problem of data imbalance.
  • the ratio of the pixels of the dental nerve canal in the obtained sampling image is increased, the influence of the data imbalance problem on the recognition accuracy is reduced, and the identified dental nerve canal is improved. accuracy.
  • FIG. 1B shows a flowchart of an algorithm for automatic segmentation of the mandibular dental neural canal according to an embodiment of the present disclosure. As shown in FIG. 1B , the method includes:
  • step S13 input the original CBCT image
  • step S14 preprocessing is performed on the output original CBCT image to obtain a preprocessed CBCT image
  • step S15 the teeth and the neural canal entrance are segmented based on the preprocessed CBCT image to obtain the segmentation result of the teeth and the dental neural canal entrance;
  • a neural network model can be used to segment the teeth in the CBCT images as well as the entrance of the dental nerve canal.
  • step S16 according to the segmentation results of the teeth and the entrance of the dental nerve canal, local rotation and resampling are performed to obtain two resampling images;
  • the regions corresponding to the teeth on both sides can be rotated and resampled according to the segmentation results of the teeth and the entrance of the tooth neural canal, to obtain two images of the corresponding regions after resampling, and the direction of the neural canal in the resampled images. Roughly parallel to the sagittal plane of the image.
  • step S17 segment the dental nerve canal in the two resampling images respectively, to obtain the segmentation result of the dental nerve canal;
  • the neural network model can be used to segment the dental neural canal in the two resampling images, respectively, to obtain the segmentation result of the dental neural canal;
  • step S18 a fast-moving minimum path extraction algorithm is used to optimize the segmentation result of the dental nerve canal to obtain an optimized segmentation result
  • the fast-moving minimum path extraction algorithm can be used to optimize the segmentation result of the dental neural canal, and repair the fracture phenomenon in the segmentation result of the dental neural canal.
  • step S19 according to the spatial coordinate information of the CBCT image, the optimized segmentation result is matched with the same physical space of the CBCT image;
  • the optimized segmentation result can be restored to the same physical space as the input CBCT according to the spatial coordinate information of the CBCT image.
  • one or more implementation manners provided by the present disclosure may be used to realize the identification of the dental neural canal. Please refer to the foregoing description for the identification process.
  • the dental neural canal is generally incomplete and discontinuous, and it is difficult to determine the exact location of the dental neural canal.
  • the position of the dental neural canal can be accurately and efficiently determined, and the obtained dental neural canal is continuous, so as to assist the doctor in performing dental implants Into the location planning, no need for doctors to invest more manpower.
  • the image processing method provided by the present disclosure may be implemented by a neural network, for example, a convolutional neural network, a recurrent neural network, and the like.
  • a neural network for example, a convolutional neural network, a recurrent neural network, and the like.
  • the neural network can output the position of the target object in the 3D image. Recognizing target objects through neural networks can improve the efficiency of target object recognition and save a lot of time.
  • the image processing method may be executed by an electronic device such as a terminal device or a server
  • the terminal device may be a user equipment (User Equipment, UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless Telephone, personal digital assistant (Personal Digital Assistant, PDA), handheld device, computing device, vehicle-mounted device, wearable device, etc.
  • the method can be implemented by the processor calling the computer-readable instructions stored in the memory.
  • the method may be performed by a server.
  • the present disclosure also provides image processing apparatuses, electronic devices, computer-readable storage media, and programs, all of which can be used to implement any image processing method provided by the present disclosure.
  • image processing apparatuses electronic devices, computer-readable storage media, and programs, all of which can be used to implement any image processing method provided by the present disclosure.
  • FIG. 2 shows a block diagram of an image processing apparatus according to an embodiment of the present disclosure.
  • the apparatus 20 includes:
  • the identification module 201 is configured to identify key features and extension directions of the target object in the three-dimensional image
  • the position determination module 202 is configured to identify the target object in the three-dimensional image according to the key feature and the extension direction, and determine the position of the target object in the three-dimensional image.
  • the identification module is configured to identify the extension direction of the reference object of the target object, and use the extension direction of the reference object of the target object as the extension direction of the target object, the The size of the reference object is larger than the size of the target object.
  • the position determination module 202 includes a resampling submodule and a first position identification submodule, and the resampling submodule is configured to extending direction, re-sampling the target area containing the target object to obtain a re-sampled image, and the re-sampled sampling plane is parallel to the extending direction of the target object;
  • the first position recognition sub-module is configured to perform target object recognition on the resampled image, and determine the position of the target object in the three-dimensional image.
  • the resampling sub-module is configured to rotate the three-dimensional image, and the extension direction of the target object in the rotated three-dimensional image is parallel to the sampling plane; according to the sampling plane, and resampling the target area of the rotated three-dimensional image to obtain a resampling image.
  • the plane where the target object is located in the rotated three-dimensional image coincides with the sampling plane
  • the resampling sub-module is configured to resample the target area of the rotated three-dimensional image along a direction perpendicular to the sampling plane to obtain a A 3D resampled image of the target object.
  • the apparatus further includes:
  • mapping relationship determination module configured to determine a mapping relationship between the resampled image and the spatial coordinates of the three-dimensional image according to the rotation angle of the three-dimensional image
  • the first position recognition sub-module is configured to perform semantic segmentation on the target object in the resampling image to obtain position information of the target object in the resampling image;
  • the position information in the sampled image and the mapping relationship determine the coordinates of the target object in the three-dimensional image.
  • the target object is in the shape of a line
  • the first position recognition sub-module is configured to, in the case that the line-shaped features of the recognized target object are discontinuous, in the three-dimensional image determining a path connecting the linear features; determining a target path in the path that satisfies a preset condition; taking the position where the linear feature and the target path are located as the position of the target object.
  • the target object includes a dental neural canal
  • the reference object of the target object includes at least two teeth adjacent to the dental neural canal
  • the three-dimensional images include cone beam computed tomography images CBCT.
  • the first position recognition sub-module is configured to The path with the lowest value of the cumulative energy function in the path is used as the target path between two adjacent linear features; the cumulative energy function is related to at least one of the following factors: the sum of the gray values of the pixels on the path, the path length, the smoothness of the path; the value of the cumulative energy function is positively correlated with the sum of the grayscale values of the pixels on the path, positively correlated with the path length, and negatively correlated with the smoothness.
  • the target area includes a to-be-implanted area in the oral cavity and an area where teeth adjacent to the to-be-implanted area are located.
  • the functions or modules included in the apparatus provided in the embodiments of the present disclosure may be configured to execute the methods described in the above method embodiments, and for implementation, reference may be made to the above method embodiments.
  • Embodiments of the present disclosure further provide a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the foregoing method is implemented.
  • the computer-readable storage medium may be a non-volatile computer-readable storage medium.
  • An embodiment of the present disclosure further provides an electronic device, comprising: a processor; a memory configured to store instructions executable by the processor; wherein the processor is configured to invoke the instructions stored in the memory to execute the above method.
  • Embodiments of the present disclosure also provide a computer program product, including computer-readable codes.
  • a processor in the device executes a method configured to implement the image processing method provided in any of the above embodiments. instruction.
  • Embodiments of the present disclosure further provide another computer program product configured to store computer-readable instructions, which, when executed, cause the computer to perform the operations of the image processing method provided by any of the foregoing embodiments.
  • the electronic device may be provided as a terminal, server or other form of device.
  • FIG. 3 shows a block diagram of an electronic device 800 according to an embodiment of the present disclosure.
  • electronic device 800 may be a mobile phone, computer, digital broadcast terminal, messaging device, game console, tablet device, medical device, fitness device, personal digital assistant, etc. terminal.
  • electronic device 800 may include one or more of the following components: processing component 802, memory 804, power supply component 806, multimedia component 808, audio component 810, input/output (I/O) interface 812, sensor component 814 , and the communication component 816 .
  • the processing component 802 generally controls the overall operation of the electronic device 800, such as operations associated with display, phone calls, data communications, camera operations, and recording operations.
  • the processing component 802 can include one or more processors 820 to execute instructions to perform all or some of the steps of the methods described above.
  • processing component 802 may include one or more modules that facilitate interaction between processing component 802 and other components.
  • processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.
  • Memory 804 is configured to store various types of data to support operation at electronic device 800 . Examples of such data include instructions for any application or method configured to operate on electronic device 800, contact data, phonebook data, messages, pictures, videos, and the like. Memory 804 may be implemented by any type of volatile or nonvolatile storage device or 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 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 or Optical Disk Magnetic Disk
  • Power supply assembly 806 provides power to various components of electronic device 800 .
  • Power supply components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800 .
  • 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 a user.
  • the touch panel includes one or more touch sensors to sense touch, swipe, and gestures on the touch panel. The touch sensor may not only sense the boundaries of a touch or swipe action, but also detect the duration and pressure associated with the touch or swipe action.
  • the multimedia component 808 includes a front-facing camera and/or a rear-facing 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 may receive external multimedia data. Each of the front and rear cameras can be a fixed optical lens system or have focal length and optical zoom capability.
  • Audio component 810 is configured to output and/or input audio signals.
  • the audio component 810 includes a microphone (MIC) that is configured to receive external audio signals when the electronic device 800 is in an operating mode, such as a calling mode, a recording mode, and a voice recognition mode.
  • the received audio signal may be further stored in memory 804 or transmitted via communication component 816 .
  • audio component 810 also includes a speaker configured to output audio signals.
  • the I/O interface 812 provides an interface between the processing component 802 and a peripheral interface module, which may be a keyboard, a click wheel, a button, or the like. These buttons may include, but are not limited to: home button, volume buttons, start button, and lock button.
  • Sensor assembly 814 includes one or more sensors configured to provide status assessment of various aspects of electronic device 800 .
  • the sensor assembly 814 can detect the on/off state of the electronic device 800, the relative positioning of the components, such as the display and the keypad of the electronic device 800, the sensor assembly 814 can also detect the electronic device 800 or one of the electronic device 800 Changes in the position of components, presence or absence of user contact with the electronic device 800 , orientation or acceleration/deceleration of the electronic device 800 and changes in the temperature of the electronic device 800 .
  • Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact.
  • Sensor assembly 814 may also include a light sensor, such as a Complementary Metal-Oxide-Semiconductor (CMOS) or Charge Coupled Device (CCD) image sensor, configured for use in imaging applications.
  • CMOS Complementary Metal-Oxide-Semiconductor
  • CCD Charge Coupled Device
  • the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
  • Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices.
  • Electronic device 800 may access wireless networks based on communication standards, such as WiFi, 2G or 3G, or a combination thereof.
  • the communication component 816 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel.
  • the communication component 816 also includes a Near Field Communication (NFC) module to facilitate short-range communication.
  • NFC Near Field Communication
  • the NFC module may be based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wide Band (UWB) technology, Bluetooth (Blue Tooth, BT) technology and other technologies to achieve.
  • RFID Radio Frequency Identification
  • IrDA Infrared Data Association
  • UWB Ultra Wide Band
  • Bluetooth Bluetooth
  • the electronic device 800 may be implemented by one or more Application Specific Integrated Circuit (ASIC), Digital Signal Process (DSP), Digital Signal Processing Device (Digital Signal Process Device) , DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic component implementation, configured to perform the above method.
  • ASIC Application Specific Integrated Circuit
  • DSP Digital Signal Process
  • DSPD Digital Signal Processing Device
  • PLD Programmable Logic Device
  • FPGA Field Programmable Gate Array
  • controller microcontroller, microprocessor or other electronic component implementation, configured to perform the above method.
  • a non-volatile computer-readable storage medium such as a memory 804 comprising computer program instructions executable by the processor 820 of the electronic device 800 to perform the above method is also provided.
  • FIG. 4 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.
  • electronic device 1900 includes processing component 1922, which in some embodiments of the present disclosure includes one or more processors, and a memory resource represented by memory 1932 configured to store instructions executable by processing component 1922 , such as applications.
  • An application program stored in 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 assembly 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 .
  • Electronic device 1900 may operate based on an operating system stored in memory 1932, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM or the like.
  • a non-volatile computer-readable storage medium such as memory 1932 comprising computer program instructions executable by processing component 1922 of electronic device 1900 to perform the above-described 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 having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the present disclosure.
  • a computer-readable storage medium may be a tangible device that can hold and store instructions for use by the instruction execution device, and may be a volatile storage medium or a non-volatile storage medium.
  • 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.
  • Examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable Programmable Read Only Memory (Electrical Programmable Read Only Memory, EPROM or Flash), Static Random Access Memory (Static Random-Access Memory, SRAM), Portable Compact Disc Read-Only Memory (CD-ROM), Digital Video Discs (DVDs), memory sticks, floppy disks, mechanical coding devices, such as punch cards or raised structures in grooves on which instructions are stored, and any suitable combination of the above.
  • RAM random access memory
  • ROM read-only memory
  • EPROM or Flash erasable Programmable Read Only Memory
  • Static Random Access Memory SRAM
  • Portable Compact Disc Read-Only Memory CD-ROM
  • DVDs Digital Video Discs
  • memory sticks floppy disks
  • mechanical coding devices such as punch cards or raised structures in grooves on which instructions are stored, and any suitable combination of the above.
  • Computer-readable storage media are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (eg, light pulses through fiber optic cables), or through electrical wires transmitted electrical signals.
  • the computer readable program instructions described herein may be downloaded to various computing/processing devices from a computer readable storage medium, or to an external computer or external storage device over 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, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and/or edge servers.
  • a network adapter card or network interface in each computing/processing device receives computer-readable program instructions from a network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing/processing device .
  • Computer program instructions for carrying out operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or instructions in one or more programming languages.
  • Source or object code written in any combination, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as the "C" language or similar programming languages.
  • the computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server implement.
  • the remote computer may 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, may be connected to an external computer (eg, use an internet service provider to connect via the internet).
  • electronic circuits such as programmable logic circuits, Field Programmable Gate Arrays (FPGA), or Programmable Logic Arrays (Programmable Logic Arrays), are personalized by utilizing state information of computer readable program instructions Array, PLA), the electronic circuitry can execute computer-readable program instructions to implement various aspects of the present disclosure.
  • These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer or other programmable data processing apparatus to produce a machine that causes the instructions when executed by the processor of the computer or other programmable data processing apparatus , resulting in means for implementing the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.
  • These computer readable program instructions can also be stored in a computer readable storage medium, these instructions cause a computer, programmable data processing apparatus and/or other equipment to operate in a specific manner, so that the computer readable medium on which the instructions are stored includes An article of manufacture comprising instructions for implementing various aspects of the functions/acts specified in one or more blocks of the flowchart and/or block diagrams.
  • Computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other equipment to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other equipment to produce a computer-implemented process , thereby causing instructions executing on a computer, other programmable data processing apparatus, or other device to implement the functions/acts specified in one or more blocks of the flowcharts and/or block diagrams.
  • each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more functions configured to implement the specified logical function(s) executable instructions.
  • the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations can be implemented in dedicated hardware-based systems that perform the specified functions or actions , or can be implemented in a combination of dedicated hardware and computer instructions.
  • the computer program product can be implemented in hardware, software or a combination thereof.
  • the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (Software Development Kit, SDK) and the like.
  • the target object in the three-dimensional image can be identified according to the extension direction in the three-dimensional image, Since the target object exists in the extension direction, other regions other than the extension direction may not be recognized, which can improve the efficiency of target object recognition.

Landscapes

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

Abstract

一种图像处理方法、装置、电子设备、存储介质和计算机程序产品，所述方法包括：识别三维图像中目标对象的关键特征和延伸方向；根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，确定所述目标对象在所述三维图像中的位置。

Description

图像处理方法及装置、电子设备、存储介质和计算机程序产品
相关申请的交叉引用
本公开基于申请号为202010733471.9、申请日为2020年7月27日、申请名称为“图像处理方法及装置、电子设备和存储介质”的中国专利申请提出，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以引入方式并入本公开。
技术领域
本公开涉及计算机技术领域，尤其涉及一种图像处理方法、装置、电子设备、存储介质和计算机程序产品。
背景技术
随着神经网络技术的发展，基于图像的目标识别技术得到了迅速发展。基于神经网络的目标识别方法可以利用神经网络的自主学习特性，提取图像特征，识别图像中的目标对象。
在相关技术中，对三维图像中目标对象的识别精度有待提高。
发明内容
本公开提出了一种图像处理技术方案。
根据本公开的一方面，提供了一种图像处理方法，包括：
识别三维图像中目标对象的关键特征和延伸方向；
根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，确定所述目标对象在所述三维图像中的位置。
在一种可能的实现方式中，所述识别三维图像中目标对象的关键特征和延伸方向，包括：
识别所述目标对象的参照物的延伸方向，将所述目标对象的参照物的延伸方向作为所述目标对象的延伸方向，所述参照物的尺寸大于所述目标对象的尺寸。
这样，考虑到目标对象的尺寸较小，为提高识别延伸方向的效率，可以通过识别尺寸较大的参照物的延伸方向，来确定目标对象的延伸方向。
在一种可能的实现方式中，所述根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，包括：
根据所述三维图像中所述关键特征和所述延伸方向，对包含所述目标对象的目标区 域进行重采样，得到重采样图像，所述重采样的采样平面与目标对象的延伸方向平行；
对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置。
这样，可以对三维图像的目标区域进行重采样，得到重采样图像，而对于三维图像中与目标对象识别无关的其它区域，可以不进行重采样，这样能够提高目标对象识别时的效率。
在一种可能的实现方式中，对包含所述目标对象的目标区域进行重采样，得到重采样图像，包括：
对所述三维图像进行旋转，旋转后的三维图像中所述目标对象的延伸方向与所述采样平面平行；
按照所述采样平面，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像。
这样，在确定目标对象的延伸方向后，在保持采样器的采样平面不变的情况下，可以通过对三维图像进行旋转，使得旋转后的三维图像中目标对象的延伸方向与采样平面平行。
在一种可能的实现方式中，在所述重采样图像为二维图像的情况下，旋转后的三维图像中所述目标对象所在平面与采样平面重合；
在所述重采样图像为三维图像的情况下，按照所述采样平面，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像，包括：
沿与所述采样平面垂直的方向，对所述旋转后的三维图像的目标区域进行重采样，得到包含所述目标对象的三维重采样图像。
在所述重采样图像为二维图像的情况下，旋转后的三维图像中所述目标对象所在平面与采样平面重合，这样在重采样的情况下既可实现对目标对象的完整采样。
在所述重采样图像为三维图像的情况下，沿与所述采样平面垂直的方向，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像内目标对象像素的占比提高，减少了数据不平衡问题对识别精度的影响，提高了识别到的目标对象的精度。
在一种可能的实现方式中，在得到所述重采样图像后，还包括：
根据所述三维图像旋转的角度，确定所述重采样图像和所述三维图像的空间坐标的映射关系；
所述对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置，包括：
对所述重采样图像中的目标对象进行语义分割，得到所述目标对象在所述重采样图像中的位置信息；
根据所述目标对象在所述重采样图像中的位置信息，以及所述映射关系，确定所述目标对象在所述三维图像中的坐标。
这样，能够将旋转后的三维图像的识别结果，映射到旋转前的三维图像中，即可确定原始的三维图像中目标对象的位置，能够方便用户在原始的三维图像中查看目标对象的位置，提高了用户体验。
在一种可能的实现方式中，所述目标对象呈线状，所述对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置，包括：
在识别到的目标对象的线状特征不连续的情况下，在所述三维图像中确定连接所述线状特征的路径；
确定所述路径中满足预设条件的目标路径；
将所述线状特征和所述目标路径所在的位置作为所述目标对象的位置。
这样，在识别到的目标对象的线状特征不连续的情况下，将连接线状特征的路径中满足预设条件的目标路径，也作为目标对象的一部分，解决了识别到的目标对象不连续的问题，提高了目标对象识别的准确度。
在一种可能的实现方式中，所述目标对象包括牙神经管，所述目标对象的参照物包括与所述牙神经管相邻的至少两个牙齿；
所述三维图像包括锥束计算机断层扫描图像CBCT。
这样，能够准确快速地确定CBCT图像中牙神经管的位置。
在一种可能的实现方式中，在所述目标对象为牙神经管、所述三维图像为CBCT的情况下，所述确定所述路径中满足预设条件的目标路径，包括：
将相邻两条线性特征之间的路径中累积能量函数的值最低的路径，作为相邻两条线性特征之间的目标路径；
所述累积能量函数与下述至少一种因素相关：路径上像素点的灰度值之和、路径长度、路径的平滑程度；
所述累积能量函数的值与所述路径上像素点的灰度值之和正相关，与所述路径长度正相关，与所述平滑程度负相关。
这样，在识别到的牙神经管的线状特征不连续的情况下，将连接线状特征的路径中累积能量函数的值最低的路径，也作为目标对象的一部分，解决了识别到的目标对象不连续的问题，提高了目标对象识别的准确度。
在一种可能的实现方式中，所述目标区域包括口腔中的待种植区域以及与待种植区域相邻的牙齿所在区域。
这样，确定的目标区域通常包括待种植区域(即缺牙位置)相邻的牙齿，以及与牙齿连接的牙神经管，包括缺牙位置对应的牙床处的牙神经管。
根据本公开的一方面，提供了一种图像处理装置，包括：
识别模块，配置为识别三维图像中目标对象的关键特征和延伸方向；
位置确定模块，配置为根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，确定所述目标对象在所述三维图像中的位置。
根据本公开的一方面，提供了一种电子设备，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调用所述存储器存储的指令，以执行上述方法。
根据本公开的一方面，提供了一种计算机可读存储介质，其上存储有计算机程序指令，所述计算机程序指令被处理器执行时实现上述方法。
根据本公开的一方面，提供了一种计算机程序产品，所述计算机程序产品包括一条或多条指令，所述一条或多条指令适于由处理器加载并执行实现上述方法。
在本公开实施例中，通过识别三维图像中目标对象的关键特征和延伸方向，在确定出关键特征后，即可在三维图像中，根据该延伸方向，对三维图像中的目标对象进行识别，由于目标对象是存在于该延伸方向上的，因此，对于延伸方向以外的其它区域则可以不进行识别，能够提高目标对象识别的效率。
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，而非限制本公开。根据下面参考附图对示例性实施例的详细说明，本公开的其它特征及方面将变得清楚。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，这些附图示出了符合本公开的实施例，并与说明书一起用于说明本公开的技术方案。
图1A示出根据本公开实施例的图像处理方法的流程图；
图1B示出根据本公开实施例的一种下颌牙神经管自动分割算法的流程图；
图2示出根据本公开实施例的一种图像处理装置的框图；
图3示出根据本公开实施例的一种电子设备的框图；
图4示出根据本公开实施例的一种电子设备的框图。
实施方式
以下将参考附图详细说明本公开的各种示例性实施例、特征和方面。附图中相同的附图标记表示功能相同或相似的元件。尽管在附图中示出了实施例的各种方面，但是除非特别指出，不必按比例绘制附图。
在这里专用的词“示例性”意为“用作例子、实施例或说明性”。这里作为“示例性”所说明的任何实施例不必解释为优于或好于其它实施例。
本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中术语“至少一种”表示多种中的任意一种或多种中的至少两种的任意组合，例如，包括A、B、C中的至少一种，可以表示包括从A、B和C构成的集合 中选择的任意一个或多个元素。
另外，为了更好地说明本公开，在下文的实施方式中给出了众多的细节。本领域技术人员应当理解，没有某些细节，本公开同样可以实施。在一些实例中，对于本领域技术人员熟知的方法、手段、元件和电路未作详细描述，以便于凸显本公开的主旨。
图1A示出根据本公开实施例的图像处理方法的流程图，如图1A所示，所述图像处理方法包括：
在步骤S11中，识别三维图像中目标对象的关键特征和延伸方向。
三维图像也可称为立体图像，三维图像可以是通过立体图像采集设备对真实物体进行采集到的，例如医疗诊断中采集的人体器官的三维影像、产品的三维图像等等，也可以是通过三维绘图技术绘制的三维图像。本公开对三维图像的形式和来源不作限定。
目标对象的关键特征用于表征目标对象关键部位的特征，是目标对象的一部分，相较于目标对象其它部位的特征而言，关键特征更易于被识别到。即，目标对象的关键特征可以是目标对象中易于被识别到的部位的特征，例如可以是目标对象中尺寸大于设定尺寸阈值的部位的特征。
举例来说，以人体口腔中的牙神经管为例，牙神经管的起始位置往往较粗，而牙神经管的末梢位置往往较细，因此，可以以牙神经管的起始位置的特征作为目标对象的关键特征。
关键特征的识别可以通过训练好的网络来进行识别，识别方式本公开对此不做限定。由于目标对象的关键特征较其它部位的尺寸更大，因此易于识别，那么，在对目标对象进行识别的过程中，可以通过复杂度较低的图像处理算法进行识别，以提高识别效率。
目标对象的延伸方向，可以是目标对象以所述关键特征为起点在三维图像中延伸的方向，由于目标对象可能只朝着一个方向延伸，也可能朝着多个方向延伸，因此该方向可以是单向的，也可以是多向的，也可以是以关键特征为中心点向周围延伸的某一平面。
后文会结合的实施方式对目标对象延伸方向的识别过程进行详细描述，此处不做赘述。
在步骤S12中，根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，确定所述目标对象在所述三维图像中的位置。
在确定出关键特征后，即可在三维图像中，沿着关键特征的延伸方向，对三维图像中的目标对象进行识别，由于目标对象是存在于该延伸方向上的，因此，对于延伸方向以外的其它区域则可以不进行识别，能够大幅提高目标对象识别的效率。
本公开提供的图像处理方法还可以有多种实现方式，在一种可能的实现方式中，所述识别三维图像中目标对象的关键特征和延伸方向，包括：识别所述目标对象的参照物的延伸方向，将所述目标对象的参照物的延伸方向作为所述目标对象的延伸方向，所述 参照物的尺寸大于所述目标对象的尺寸。
考虑到目标对象的尺寸较小，为提高识别延伸方向的效率，可以通过识别尺寸较大的参照物的延伸方向，来确定目标对象的延伸方向。该参照物为预先选定的、延伸方向与目标对象的延伸方向相同的事物。
例如，在目标对象为牙神经管的情况下，目标对象的参照物可以是与牙神经管相邻的至少两个牙齿，由于与牙神经管相邻的至少两个牙齿的延伸方向与牙神经管的延伸方向相同，因此，可以将与牙神经管相邻的至少两个牙齿作为目标对象的参照物。并且，由于牙齿的尺寸较大，因此可以很方便地确定两个牙齿的延伸方向，提高识别延伸方向的效率。
另外，本公开还提供一种识别目标对象延伸方向的实现方式，在确定目标对象延伸方向的过程中，可以通过对三维图像中的目标对象进行粗略的识别，这样虽然不能得到目标对象的全貌，但是能得到目标对象的多个目标特征，然后将多个目标特征连接起来构成的延伸方向，确定为目标对象的延伸方向。
例如，对于处于某一平面上的线状目标对象，在进行粗略地识别后，往往会得到目标对象的多个不连续的特征点，多个特征点所在的面的延伸方向，即为目标对象的延伸方向；或者对于面状的目标对象，在进行粗略地识别后，往往会得到目标对象的多个不连续的特征线，将多个特征线连接起来的面所在的方向，即为目标对象的延伸方向。
在本公开实施例中，在粗略识别过程中，可以通过低于设定阈值的采样率对三维图像进行下采样，对下采样后得到的图像进行目标对象识别，由于采样率较低，因此，虽然不能得到目标对象的全貌，但是能根据得到的多个目标特征，来确定目标对象的延伸方向。通过该实现方式可以快速地确定目标对象的大体位置。
在一种可能的实现方式中，所述根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，包括：根据所述三维图像中所述关键特征和所述延伸方向，对包含所述目标对象的目标区域进行重采样，得到重采样图像，所述重采样的采样平面与目标对象的延伸方向平行；对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置。
在确定了三维图像中目标对象的关键特征和所述延伸方向后，以关键特征为起点的延伸方向上，即为包含目标对象的目标区域，因此，在进行目标对象的识别的情况下，可以对目标区域进行识别。
可以对三维图像的目标区域进行重采样(resample)，得到重采样图像，而对于三维图像中与目标对象识别无关的其它区域，可以不进行重采样，这样能够提高目标对象识别时的效率。
在进行重采样的过程中，可以保持采样平面与目标对象的延伸方向平行。采样平面 可以是对图像进行采样的过程中，单次采集的一张二维图像所在的平面，例如，对于xyz坐标系下的三维图像而言，采样平面可以是三维图像的yz面，而在目标对象的延伸方向也是在yz面中的情况下，可以将目标对象采集到一张二维图像中。
在进行目标对象识别的过程中，可以使用图像识别技术、图像语义分割技术、目标检测技术等图像处理技术，来确定目标对象在重采样图像中的位置，由于重采样图像是根据三维图像重采样得到，因此，可以建立重采样图像与三维图形的空间映射关系，进而根据重采样图像的识别结果确定目标对象在三维图像中的位置。
在本公开实施例中，通过确定三维图像中包含待识别目标对象的目标区域，然后在对目标区域进行重采样的过程中，保持采样平面与目标对象的延伸方向平行，得到重采样图像；再对重采样图像进行目标对象识别，确定目标对象在三维图像中的位置。通过使采样平面与目标对象的延伸方向平行，可以尽可能地让目标对象被采集到一整个重采样图像内，这样便可以对重采样图像内完整的目标对象进行分割，另外，在对三维图像旋转后，得到重采样图像内目标对象像素的占比提高，减少了数据不平衡问题对识别精度的影响，提高识别到的目标对象的精度。
在一种可能的实现方式中，对所述目标区域进行重采样，得到重采样图像，包括：对所述三维图像进行旋转，旋转后的三维图像中所述目标对象的延伸方向与采样平面平行；按照所述采样平面，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像。
在确定目标对象的延伸方向后，在保持采样器的采样平面不变的情况下，可以通过对三维图像进行旋转，使得旋转后的三维图像中目标对象的延伸方向与采样平面平行。
由于重采样图像是在采样平面与目标对象的延伸方向平行的情况下采集得到的，因此，在重采样图像中往往能完整地包含目标对象。在该情况下，对重采样图像进行语义分割，便能够完整地分割出目标对象，提高目标对象的分割精度。
所述重采样图像可以是二维图像，也可以是三维图像，即对所述目标区域进行重采样的过程中，即可以是采集二维图像，也可以是采集三维图像。
在一种可能的实现方式中，在所述重采样图像为二维图像的情况下，旋转后的三维图像中所述目标对象所在平面与采样平面重合，这样在重采样的情况下既可实现对目标对象的完整采样。如上文所述，对于三维图像中二维的目标对象，例如，线状、面状的目标对象，可先确定目标对象所在平面，在确定目标对象所在平面后，并以此为采样平面对图形进行重采样。
由于重采样图像是在采样平面与目标对象所在平面重合的情况下采集得到的，因此，在重采样图像中往往能完整地包含目标对象。在该情况下，对重采样图像进行目标识别，便能够完整地识别出目标对象，提高识别到的目标对象的精度，减少在目标对象 与采样平面不平行的情况下，目标对象识别结果较差的情况。
在一种可能的实现方式中，在所述重采样图像为三维图像的情况下，按照所述采样平面，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像，包括：沿与采样平面垂直的方向，对所述旋转后的三维图像的目标区域进行重采样，得到包含所述目标对象的三维重采样图像。
对于三维的目标对象，在三维空间中存在xyz三个方向的尺寸，这三个方向的尺寸中，两个较长的尺寸所在的方向所构成的平面，即为三维的目标对象的延伸方向。
在确定目标对象所在平面后，在保持采样器的采样平面不变的情况下，可以通过对三维图像进行旋转，使得旋转后的三维图像中目标对象的延伸方向与采样平面平行，在对三维图像旋转后，得到重采样图像内目标对象像素的占比提高，减少了数据不平衡问题对识别精度的影响，提高了识别到的目标对象的精度。
在一种可能的实现方式中，在得到所述重采样图像后，还包括：根据所述三维图像旋转的角度，确定所述重采样图像和所述三维图像的空间坐标的映射关系；所述对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置，包括：对所述重采样图像中的目标对象进行语义分割，得到所述目标对象在所述重采样图像中的位置信息；根据所述目标对象在所述重采样图像中的位置信息，以及所述映射关系，确定所述目标对象在所述三维图像中的坐标。
语义分割是在图像的像素级别上的分类，图像中属于同一类的像素会被归为一类，通过对重采样图像中的目标对象进行语义分割，将会得到重采样图像中属于目标对象这一类的像素，目标对象的像素的位置信息即为重采样图像中目标对象的位置信息。进行语义分割的过程可通过训练好的网络来实现，该网络可通过带有标记的样本进行训练得到，本公开对此不做赘述。
在对三维图像进行旋转的过程中，可以以三维空间中的某一点为旋转中心，对三维图像进行旋转，并确定三维图像相对于旋转中心旋转的角度。然后根据三维图像旋转的角度，确定采样图像和三维图像的空间坐标的映射关系。该映射关系可以是三维图像中的像素点，在旋转前后坐标的对应关系。
由于对三维图像进行了旋转，因此三维图像在空间坐标系中的坐标发生了变化，在本公开实施例中，能够将旋转后的三维图像的识别结果，映射到旋转前的三维图像中，即可确定原始的三维图像中目标对象的位置，能够方便用户在原始的三维图像中查看目标对象的位置，提高了用户体验。
在一种可能的实现方式中，所述目标对象呈线状，所述对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置，包括：在识别到的目标对象的 线状特征不连续的情况下，在所述三维图像中确定连接所述线状特征的路径；确定所述路径中满足预设条件的目标路径；将所述线状特征和所述目标路径所在的位置作为所述目标对象的位置。
在图像识别技术中，往往会存在目标对象的局部在图像上存在但是未识别出的情况，或者目标对象的局部在图像清晰度不够，不足以识别，甚至不可见的情况，在本公开中，在识别到的目标对象的线状特征不连续的情况下，可以确定出连接已识别出的线状特征的路径，这些路径中便会存在未识别出的线状特征。因此，可以从这些路径中选择满足预设条件的目标路径，作为目标识别过程中未识别出的线状特征。
在本公开实施例中，在识别到的目标对象的线状特征不连续的情况下，将连接线状特征的路径中满足预设条件的目标路径，也作为目标对象的一部分，解决了识别到的目标对象不连续的问题，提高了目标对象识别的准确度。
在一种可能的实现方式中，所述目标对象包括牙神经管；所述三维图像包括锥束计算机断层扫描图像(Cone beam computed tomography，CBCT)。
在一种可能的实现方式中，在所述目标对象为牙神经管、所述三维图像为CBCT的情况下，预设条件的目标路径，可以是相邻两条线性特征之间的路径中的累积能量函数的值最低的路径，那么，确定所述路径中满足预设条件的目标路径，可以包括：将相邻两条线性特征之间的路径中累积能量函数的值最低的路径，作为相邻两条线性特征之间的目标路径。
所述累积能量函数与下述至少一种因素相关：路径上像素点的灰度值之和；路径长度；路径的平滑程度；
所述累积能量函数的值与所述路径上像素点的灰度值之和正相关，与所述路径长度正相关，与所述平滑程度负相关。
在CBCT图像中，考虑到牙神经管的灰度值往往低于下颌骨等其它组织器官的灰度值，因此，路径像素点的灰度值之和越低，则累积能量函数的值越低，则路径为牙神经管的概率越大。
相对于牙神经管附近的其它组织器官而言，牙神经管的路径较为接近于直线，因此，路径越短，累积能量函数的值越低，则路径为牙神经管的概率越大。
相对于牙神经管附近的其它组织器官而言，牙神经管的平滑程度往往较高，因此，平滑程度越高，累积能量函数的值越低，则路径为牙神经管的概率越大。
在本公开实施例中，在识别到的牙神经管的线状特征不连续的情况下，将连接线状特征的路径中累积能量函数的值最低的路径，也作为目标对象的一部分，解决了识别到的目标对象不连续的问题，提高了目标对象识别的准确度。
在一种可能的实现方式中，包含待识别目标对象的目标区域，包括口腔中的待种植 区域以及与待种植区域相邻的牙齿所在区域。这样确定的目标区域通常包括待种植区域(即缺牙位置)相邻的牙齿，以及与牙齿连接的牙神经管，包括缺牙位置对应的牙床处的牙神经管。
种植牙是目前较为常见的一种缺牙修复方式。在种植牙植入手术中，种植牙的植入位置直接影响手术的成功与否。在手术规划中，种植牙的植入位置应避开位于牙床的牙神经管，避免对牙神经管造成损伤和压迫。
通过本公开实施例，能够准确快速地确定CBCT图像中牙神经管的位置，在确定牙神经管的过程中，会确定CBCT口腔图像中的待种植区域以及与待种植区域相邻的牙齿所在区域；然后对该区域进行重采样，得到重采样图像，其中通过对CBCT图像的旋转，使采样平面与牙神经管的延伸方向平行，这样可以使得牙神经管尽可能完整地被采集；然后对采样图像进行牙神经管识别，在识别出的牙神经管不连续的情况下，可利用识别出的相邻牙神经管片段之间的最短路径进行补充完整，即可确定牙神经管在重采样图像中的位置，进而根据映射关系，确定牙神经管在原CBCT图像中的位置。
另外，在CBCT图像中，往往是整个口腔及其周边区域的图像，而牙神经管所占的像素点数量较少，像素点所占的比重较低，在通过网络对牙神经管进行识别的过程中，往往会存在数据不平衡的问题。在本公开实施例中，通过对三维CBCT图像进行旋转，使得到的采样图像内牙神经管像素的占比提高，减少了数据不平衡问题对识别精度的影响，提高了识别到的牙神经管的精度。
图1B示出根据本公开实施例的一种下颌牙神经管自动分割算法的流程图，如图1B所示，所述方法包括：
在步骤S13中，输入原始CBCT图像；
在步骤S14中，对输出的原始CBCT图像做预处理，得到完成预处理的CBCT图像；
在步骤S15中，基于完成预处理的CBCT图像做牙齿及神经管入口分割，得到牙齿以及牙神经管入口的分割结果；
在实施过程中，可以使用神经网络模型对CBCT图像中的牙齿以及牙神经管入口进行分割。
在步骤S16中，根据牙齿以及牙神经管入口的分割结果，做局部旋转与重采样，得到两幅重采样图像；
在实施过程中，可以根据牙齿以及牙神经管入口的分割结果，分别对两侧牙齿对应的区域进行旋转和重采样，得到两幅重采样后对应区域的图像，重采样图像中神经管的走向与图像的矢状面大致平行。
在步骤S17中，分别对两幅重采样图像中的牙神经管进行分割，得到牙神经管的分割结果；
在实施过程中，可以使用神经网络模型，分别对两幅重采样图像中的牙神经管进行 分割，得到牙神经管的分割结果；
在步骤S18中，采用快速行进最小路径提取算法，对牙神经管的分割结果进行优化，得到优化后的分割结果；
在实施过程中，可以根据神经网络模型的预测结果，采用快速行进最小路径提取算法，对牙神经管分割结果进行优化，修补牙神经管分割结果中出现的断裂现象。
在步骤S19中，根据CBCT图像的空间坐标信息，将优化后的分割结果与CBCT图像相同的物理空间匹配；
在实施过程中，可以根据CBCT图像的空间坐标信息，将优化后的分割结果恢复到与输入CBCT相同的物理空间中。
需要说明的是，在对牙神经管识别的过程中，可以结合本公开提供的一个或多个实现方式来实现对牙神经管的识别，识别过程请参阅前文描述。
在CBCT图像中，牙神经管普遍呈现不完整、不连续，难以确定牙神经管的准确位置。在本公开实施例中，通过对三维的CBCT图像进行牙神经管的识别，能够精确、高效地确定牙神经管的位置，并且得到的牙神经管是连续的，以此辅助医生进行种植牙植入位置规划，无需医生投入较多的人力。
在一种可能的实现方式中，本公开提供的图像处理方法可通过神经网络来实现，例如，卷积神经网络、循环神经网络等等。通过将三维网络输入到训练好的神经网络中，神经网络即可输出目标对象在三维图像中的位置。通过神经网络对目标对象进行识别，能够提高目标对象识别的效率，节省大量时间。
在一种可能的实现方式中，所述图像处理方法可以由终端设备或服务器等电子设备执行，终端设备可以为用户设备(User Equipment，UE)、移动设备、用户终端、终端、蜂窝电话、无绳电话、个人数字处理(Personal Digital Assistant，PDA)、手持设备、计算设备、车载设备、可穿戴设备等，所述方法可以通过处理器调用存储器中存储的计算机可读指令的方式来实现。或者，可通过服务器执行所述方法。
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。本领域技术人员可以理解，在实施方式的上述方法中，各步骤的执行顺序应当以其功能和可能的内在逻辑确定。
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。
图2示出根据本公开实施例的图像处理装置的框图，如图2所示，所述装置20包括：
识别模块201，配置为识别三维图像中目标对象的关键特征和延伸方向；
位置确定模块202，配置为根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，确定所述目标对象在所述三维图像中的位置。
在一种可能的实现方式中，所述识别模块，配置为识别所述目标对象的参照物的延伸方向，将所述目标对象的参照物的延伸方向作为所述目标对象的延伸方向，所述参照物的尺寸大于所述目标对象的尺寸。
在一种可能的实现方式中，所述位置确定模块202包括重采样子模块和第一位置识别子模块，所述重采样子模块，配置为根据所述三维图像中所述关键特征和所述延伸方向，对包含所述目标对象的目标区域进行重采样，得到重采样图像，所述重采样的采样平面与目标对象的延伸方向平行；
所述第一位置识别子模块，配置为对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置。
在一种可能的实现方式中，所述重采样子模块，配置为对所述三维图像进行旋转，旋转后的三维图像中所述目标对象的延伸方向与所述采样平面平行；按照所述采样平面，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像。
在一种可能的实现方式中，在所述重采样图像为二维图像的情况下，旋转后的三维图像中所述目标对象所在平面与采样平面重合；
在所述重采样图像为三维图像的情况下，所述重采样子模块，配置为沿与所述采样平面垂直的方向，对所述旋转后的三维图像的目标区域进行重采样，得到包含所述目标对象的三维重采样图像。
在一种可能的实现方式中，所述装置还包括：
映射关系确定模块，配置为根据所述三维图像旋转的角度，确定所述重采样图像和所述三维图像的空间坐标的映射关系；
所述第一位置识别子模块，配置为对所述重采样图像中的目标对象进行语义分割，得到所述目标对象在所述重采样图像中的位置信息；根据所述目标对象在所述重采样图像中的位置信息，以及所述映射关系，确定所述目标对象在所述三维图像中的坐标。
在一种可能的实现方式中，所述目标对象呈线状，所述第一位置识别子模块，配置为在识别到的目标对象的线状特征不连续的情况下，在所述三维图像中确定连接所述线状特征的路径；确定所述路径中满足预设条件的目标路径；将所述线状特征和所述目标路径所在的位置作为所述目标对象的位置。
在一种可能的实现方式中，所述目标对象包括牙神经管，所述目标对象的参照物包括与所述牙神经管相邻的至少两个牙齿；
所述三维图像包括锥束计算机断层扫描图像CBCT。
在一种可能的实现方式中，在所述目标对象为牙神经管、所述三维图像为CBCT的情况下，所述第一位置识别子模块，配置为将相邻两条线性特征之间的路径中累积能量函数的值最低的路径，作为相邻两条线性特征之间的目标路径；所述累积能量函数与下述至少一种因素相关：路径上像素点的灰度值之和、路径长度、路径的平滑程度；所述累积能量函数的值与所述路径上像素点的灰度值之和正相关，与所述路径长度正相关，与所述平滑程度负相关。
在一种可能的实现方式中，所述目标区域包括口腔中的待种植区域以及与待种植区域相邻的牙齿所在区域。
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配置为执行上文方法实施例描述的方法，其实现可以参照上文方法实施例的描述。
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；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调用所述存储器存储的指令，以执行上述方法。
本公开实施例还提供了一种计算机程序产品，包括计算机可读代码，当计算机可读代码在设备上运行时，设备中的处理器执行配置为实现如上任一实施例提供的图像处理方法的指令。
本公开实施例还提供了另一种计算机程序产品，配置为存储计算机可读指令，指令被执行时使得计算机执行上述任一实施例提供的图像处理方法的操作。
电子设备可以被提供为终端、服务器或其它形态的设备。
图3示出根据本公开实施例的一种电子设备800的框图。例如，电子设备800可以是移动电话，计算机，数字广播终端，消息收发设备，游戏控制台，平板设备，医疗设备，健身设备，个人数字助理等终端。
参照图3，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配置为在电子设备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(Liquid Crystal Display，LCD)和触摸面板(Touch panel，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(microphone，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，配置为输出音频信号。
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，配置为为电子设备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，如互补金属氧化物半导体(Complementary Metal-Oxide-Semiconductor，CMOS)或电荷耦合器件(Charge Coupled Device，CCD)图像传感器，配置为在成像应用中使用。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(Near Field Communication，NFC)模块，以促进短程通信。例如，在NFC模块可基于射频识别(Radio Frequency Identification，RFID)技术，红外数据协会(Infrared Data Association，IrDA)技术，超宽带(Ultra Wide Band，UWB)技术，蓝牙(Blue Tooth，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(Application Specific Integrated Circuit，ASIC)、数字信号处理器(Digital Signal Process，DSP)、数字信号处理设备(Digital Signal Process Device，DSPD)、可编程逻辑器件(Programmable Logic Device，PLD)、现场可编程门阵列(Field Programmable Gate Array，FPGA)、控制器、微控制器、微处理器或其他电子元件实现，配置为执行上述方法。
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4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4，电子设备1900包括处理组件1922，其在本公开的一些实施例中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，配置为存储可由处理组件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，可为易失性存储介质或非易失性存储介质。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(Random Access Memory，RAM)、只读存储器 (Read-Only Memory，ROM)、可擦式可编程只读存储器(Electrical Programmable Read Only Memory，EPROM或闪存)、静态随机存取存储器(Static Random-Access Memory，SRAM)、便携式压缩盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、数字多功能盘(Digital Video Disc，DVD)、记忆棒、软盘、机械编码设备、例如其上存储有指令的打孔卡或凹槽内凸起结构、以及上述的任意合适的组合。这里所使用的计算机可读存储介质不被解释为瞬时信号本身，诸如无线电波或者其他自由传播的电磁波、通过波导或其他传输媒介传播的电磁波(例如，通过光纤电缆的光脉冲)、或者通过电线传输的电信号。
这里所描述的计算机可读程序指令可以从计算机可读存储介质下载到各个计算/处理设备，或者通过网络、例如因特网、局域网、广域网和/或无线网下载到外部计算机或外部存储设备。网络可以包括铜传输电缆、光纤传输、无线传输、路由器、防火墙、交换机、网关计算机和/或边缘服务器。每个计算/处理设备中的网络适配卡或者网络接口从网络接收计算机可读程序指令，并转发该计算机可读程序指令，以供存储在各个计算/处理设备中的计算机可读存储介质中。
用于执行本公开操作的计算机程序指令可以是汇编指令、指令集架构(ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码，所述编程语言包括面向对象的编程语言—诸如Smalltalk、C++等，以及常规的过程式编程语言—诸如“C”语言或类似的编程语言。计算机可读程序指令可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络—包括局域网(Local Area Network，LAN)或广域网(Wide Area Network，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(Field Programmable Gate Array，FPGA)或可编程逻辑阵列(Programmable Logic Array，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配置为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。
该计算机程序产品可以通过硬件、软件或其结合的方式实现。在一个可选实施例中，所述计算机程序产品体现为计算机存储介质，在另一个可选实施例中，计算机程序产品体现为软件产品，例如软件开发包(Software Development Kit，SDK)等等。
以上已经描述了本公开的各实施例，上述说明是示例性的，并非穷尽性的，并且也不限于所披露的各实施例。在不偏离所说明的各实施例的范围和精神的情况下，对于本技术领域的普通技术人员来说许多修改和变更都是显而易见的。本文中所用术语的选择，旨在最好地解释各实施例的原理、实际应用或对市场中的技术的改进，或者使本技术领域的其它普通技术人员能理解本文披露的各实施例。
工业实用性
在本公开实施例中，通过识别三维图像中目标对象的关键特征和延伸方向，在确定出关键特征后，即可在三维图像中，根据该延伸方向，对三维图像中的目标对象进行识别，由于目标对象是存在于该延伸方向上的，因此，对于延伸方向以外的其它区域则可以不进行识别，能够提高目标对象识别的效率。

Claims (23)

  1. 一种图像处理方法，包括：
    识别三维图像中目标对象的关键特征和延伸方向；
    根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，确定所述目标对象在所述三维图像中的位置。
  2. 根据权利要求1所述方法，其中，所述识别三维图像中目标对象的关键特征和延伸方向，包括：
    识别所述目标对象的参照物的延伸方向，将所述目标对象的参照物的延伸方向作为所述目标对象的延伸方向，所述参照物的尺寸大于所述目标对象的尺寸。
  3. 根据权利要求1所述方法，其中，所述根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，包括：
    根据所述三维图像中所述关键特征和所述延伸方向，对包含所述目标对象的目标区域进行重采样，得到重采样图像，所述重采样的采样平面与目标对象的延伸方向平行；
    对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置。
  4. 根据权利要求3所述方法，其中，对包含所述目标对象的目标区域进行重采样，得到重采样图像，包括：
    对所述三维图像进行旋转，旋转后的三维图像中所述目标对象的延伸方向与所述采样平面平行；
    按照所述采样平面，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像。
  5. 根据权利要求4所述方法，其中，在所述重采样图像为二维图像的情况下，旋转后的三维图像中所述目标对象所在平面与采样平面重合；
    在所述重采样图像为三维图像的情况下，按照所述采样平面，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像，包括：
    沿与所述采样平面垂直的方向，对所述旋转后的三维图像的目标区域进行重采样，得到包含所述目标对象的三维重采样图像。
  6. 根据权利要求3所述方法，其中，在得到所述重采样图像后，还包括：
    根据所述三维图像旋转的角度，确定所述重采样图像和所述三维图像的空间坐标的映射关系；
    所述对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置，包括：
    对所述重采样图像中的目标对象进行语义分割，得到所述目标对象在所述重采样图像中的位置信息；
    根据所述目标对象在所述重采样图像中的位置信息，以及所述映射关系，确定所述目标对象在所述三维图像中的坐标。
  7. 根据权利要求3所述方法，其中，所述目标对象呈线状，所述对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置，包括：
    在识别到的目标对象的线状特征不连续的情况下，在所述三维图像中确定连接所述线状特征的路径；
    确定所述路径中满足预设条件的目标路径；
    将所述线状特征和所述目标路径所在的位置作为所述目标对象的位置。
  8. 根据权利要求2至7任一所述方法，其中，所述目标对象包括牙神经管，所述目标对象的参照物包括与所述牙神经管相邻的至少两个牙齿；
    所述三维图像包括锥束计算机断层扫描图像CBCT。
  9. 根据权利要求8所述方法，其中，在所述目标对象为牙神经管、所述三维图像为CBCT的情况下，所述确定所述路径中满足预设条件的目标路径，包括：
    将相邻两条线性特征之间的路径中累积能量函数的值最低的路径，作为相邻两条线性特征之间的目标路径；
    所述累积能量函数与下述至少一种因素相关：路径上像素点的灰度值之和、路径长度、路径的平滑程度；
    所述累积能量函数的值与所述路径上像素点的灰度值之和正相关，与所述路径长度正相关，与所述平滑程度负相关。
  10. 根据权利要求9所述的方法，其中，所述目标区域包括口腔中的待种植区域以及与待种植区域相邻的牙齿所在区域。
  11. 一种图像处理装置，包括：
    识别模块，配置为识别三维图像中目标对象的关键特征和延伸方向；
    位置确定模块，配置为根据所述关键特征和所述延伸方向，在所述三维图像中对所述目标对象进行识别，确定所述目标对象在所述三维图像中的位置。
  12. 根据权利要求11所述的装置，其中，所述识别模块，配置为识别所述目标对象的参照物的延伸方向，将所述目标对象的参照物的延伸方向作为所述目标对象的延伸方向，所述参照物的尺寸大于所述目标对象的尺寸。
  13. 根据权利要求11所述的装置，其中，所述位置确定模块包括重采样子模块和第一位置识别子模块，所述重采样子模块，配置为根据所述三维图像中所述关键特征和所述延伸方向，对包含所述目标对象的目标区域进行重采样，得到重采样图像，所述重采样的采样平面与目标对象的延伸方向平行；所述第一位置识别子模块，配置为对所述重采样图像进行目标对象识别，确定所述目标对象在所述三维图像中的位置。
  14. 根据权利要求13所述的装置，其中，所述重采样子模块，配置为对所述三维图像进行旋转，旋转后的三维图像中所述目标对象的延伸方向与所述采样平面平行；按照所述采样平面，对所述旋转后的三维图像的目标区域进行重采样，得到重采样图像。
  15. 根据权利要求14所述的装置，其中，在所述重采样图像为二维图像的情况下，旋转后的三维图像中所述目标对象所在平面与采样平面重合；在所述重采样图像为三维图像的情况下，所述重采样子模块，配置为沿与所述采样平面垂直的方向，对所述旋转后的三维图像的目标区域进行重采样，得到包含所述目标对象的三维重采样图像。
  16. 根据权利要求13所述的装置，其中，所述装置还包括：映射关系确定模块，配置为根据所述三维图像旋转的角度，确定所述重采样图像和所述三维图像的空间坐标的映射关系；所述第一位置识别子模块，配置为对所述重采样图像中的目标对象进行语义分割，得到所述目标对象在所述重采样图像中的位置信息；根据所述目标对象在所述重采样图像中的位置信息，以及所述映射关系，确定所述目标对象在所述三维图像中的坐标。
  17. 根据权利要求13所述的装置，其中，所述目标对象呈线状，所述第一位置识别子模块，配置为在识别到的目标对象的线状特征不连续的情况下，在所述三维图像中确定连接所述线状特征的路径；确定所述路径中满足预设条件的目标路径；将所述线状特征和所述目标路径所在的位置作为所述目标对象的位置。
  18. 根据权利要求12至17任一所述的装置，其中，所述目标对象包括牙神经管，所述目标对象的参照物包括与所述牙神经管相邻的至少两个牙齿；所述三维图像包括锥束计算机断层扫描图像CBCT。
  19. 根据权利要求18所述的装置，其中，在所述目标对象为牙神经管、所述三维图像为CBCT的情况下，所述第一位置识别子模块，配置为将相邻两条线性特征之间的路径中累积能量函数的值最低的路径，作为相邻两条线性特征之间的目标路径；所述累积能量函数与下述至少一种因素相关：路径上像素点的灰度值之和、路径长度、路径的平滑程度；所述累积能量函数的值与所述路径上像素点的灰度值之和正相关，与所述路径长度正相关，与所述平滑程度负相关。
  20. 根据权利要求19所述的装置，其中，所述目标区域包括口腔中的待种植区域以及与待种植区域相邻的牙齿所在区域。
  21. 一种电子设备，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调用所述存储器存储的指令，以执行权利要求1至10中任意一项所述的方法。
  22. 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权利要求1至10中任意一项所述的方法。
  23. 一种计算机程序产品，所述计算机程序产品包括一条或多条指令，所述一条或多条指令适于由处理器加载并执行如权利要求1至10任一项所述的图像处理方法。
PCT/CN2021/107512 2020-07-27 2021-07-21 图像处理方法及装置、电子设备、存储介质和计算机程序产品 WO2022022350A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010733471.9 2020-07-27
CN202010733471.9A CN111860388A (zh) 2020-07-27 2020-07-27 图像处理方法及装置、电子设备和存储介质

Publications (1)

Publication Number Publication Date
WO2022022350A1 true WO2022022350A1 (zh) 2022-02-03

Family

ID=72947881

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/107512 WO2022022350A1 (zh) 2020-07-27 2021-07-21 图像处理方法及装置、电子设备、存储介质和计算机程序产品

Country Status (2)

Country Link
CN (1) CN111860388A (zh)
WO (1) WO2022022350A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117857912A (zh) * 2024-01-15 2024-04-09 中山大学附属口腔医院 用于牙科的信息采集方法和系统

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111860388A (zh) * 2020-07-27 2020-10-30 上海商汤智能科技有限公司 图像处理方法及装置、电子设备和存储介质
CN112397198B (zh) * 2020-11-17 2024-08-16 上海商汤善萃医疗科技有限公司 图像处理方法及装置、电子设备和存储介质
CN112907517B (zh) * 2021-01-28 2024-07-19 上海商汤善萃医疗科技有限公司 一种图像处理方法、装置、计算机设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108364356A (zh) * 2018-02-28 2018-08-03 辽宁科技大学 一种牙齿三维网格模型的自动分割方法
CN108470375A (zh) * 2018-04-26 2018-08-31 重庆市劢齐医疗科技有限责任公司 基于深度学习的神经导管自动检测算法
CN109598703A (zh) * 2018-11-19 2019-04-09 影为医疗科技(上海)有限公司 牙齿图像的处理方法、系统、计算机可读存储介质及设备
CN111241887A (zh) * 2018-11-29 2020-06-05 北京市商汤科技开发有限公司 目标对象关键点识别方法及装置、电子设备和存储介质
CN111860388A (zh) * 2020-07-27 2020-10-30 上海商汤智能科技有限公司 图像处理方法及装置、电子设备和存储介质

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11618438B2 (en) * 2018-03-26 2023-04-04 International Business Machines Corporation Three-dimensional object localization for obstacle avoidance using one-shot convolutional neural network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108364356A (zh) * 2018-02-28 2018-08-03 辽宁科技大学 一种牙齿三维网格模型的自动分割方法
CN108470375A (zh) * 2018-04-26 2018-08-31 重庆市劢齐医疗科技有限责任公司 基于深度学习的神经导管自动检测算法
CN109598703A (zh) * 2018-11-19 2019-04-09 影为医疗科技(上海)有限公司 牙齿图像的处理方法、系统、计算机可读存储介质及设备
CN111241887A (zh) * 2018-11-29 2020-06-05 北京市商汤科技开发有限公司 目标对象关键点识别方法及装置、电子设备和存储介质
CN111860388A (zh) * 2020-07-27 2020-10-30 上海商汤智能科技有限公司 图像处理方法及装置、电子设备和存储介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117857912A (zh) * 2024-01-15 2024-04-09 中山大学附属口腔医院 用于牙科的信息采集方法和系统
CN117857912B (zh) * 2024-01-15 2024-06-07 中山大学附属口腔医院 用于牙科的信息采集方法和系统

Also Published As

Publication number Publication date
CN111860388A (zh) 2020-10-30

Similar Documents

Publication Publication Date Title
WO2022022350A1 (zh) 图像处理方法及装置、电子设备、存储介质和计算机程序产品
WO2022151755A1 (zh) 目标检测方法及装置、电子设备、存储介质、计算机程序产品和计算机程序
US20210319560A1 (en) Image processing method and apparatus, and storage medium
WO2022036972A1 (zh) 图像分割方法及装置、电子设备和存储介质
WO2022134382A1 (zh) 图像分割方法及装置、电子设备和存储介质、计算机程序
WO2020007241A1 (zh) 图像处理方法和装置、电子设备以及计算机可读存储介质
US20210097715A1 (en) Image generation method and device, electronic device and storage medium
US20210319538A1 (en) Image processing method and device, electronic equipment and storage medium
TW202110387A (zh) 圖像處理方法、電子設備、電腦可讀儲存介質
WO2020181728A1 (zh) 图像处理方法及装置、电子设备和存储介质
TWI718631B (zh) 人臉圖像的處理方法及裝置、電子設備和儲存介質
WO2023050691A1 (zh) 图像处理方法及装置、电子设备、存储介质和程序
WO2022142298A1 (zh) 关键点检测方法及装置、电子设备和存储介质
TWI765386B (zh) 神經網路訓練及圖像分割方法、電子設備和電腦儲存介質
WO2022134475A1 (zh) 点云地图构建方法及装置、电子设备、存储介质和程序
JP2022548453A (ja) 画像分割方法及び装置、電子デバイス並びに記憶媒体
WO2022193456A1 (zh) 目标跟踪方法及装置、电子设备和存储介质
CN112184787A (zh) 图像配准方法及装置、电子设备和存储介质
WO2023273498A1 (zh) 深度检测方法及装置、电子设备和存储介质
WO2023273499A1 (zh) 深度检测方法及装置、电子设备和存储介质
WO2022100005A1 (zh) 牙齿图像的处理方法、装置、电子设备、存储介质及程序
CN112613447B (zh) 关键点检测方法及装置、电子设备和存储介质
US11403788B2 (en) Image processing method and apparatus, electronic device, and storage medium
CN113660531A (zh) 视频处理方法及装置、电子设备和存储介质
CN112767347A (zh) 一种图像配准方法及装置、电子设备和存储介质

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: 21850414

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

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

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 11.07.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21850414

Country of ref document: EP

Kind code of ref document: A1