Image snapshot method and device, shooting equipment and vehicle

Abstract

The invention provides an image snapshot method and device, shooting equipment and a vehicle, and relates to the technical field of electronic equipment. According to the method, real-time processing is carried out through the images shot in the advancing process of the user shooting equipment in combination with the personalized shooting preference of the user, preference information at least comes from historical photos and / or historical videos of the user, and then a plurality of images meeting the snapshot expectation of the user are actively captured from the images to serve as candidates to be stored; and subsequently, a snapshot triggering instruction of the user can be monitored, a plurality of images can be extracted from the snapshot triggering instruction to serve as target snapshot images with original definition, and the user can also select the target snapshot images from all the images by himself / herself. According to the method, the problem that the snapshot object is missed due to advancing and response delay is effectively solved, and the data scale can be greatly reduced through the capturing link, so that the method can adapt to the existing storage space without executing compression processing, and the definition of the actually reserved target snapshot image is further ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic equipment, and in particular to an image capturing method, device, shooting equipment and vehicle. Background Art

[0002] It is common to equip vehicles with image capture equipment, such as but not limited to several cameras installed on vehicles such as cars, which can provide functions such as driving recording, intelligent driving, and sentry mode.

[0003] As for the function of taking snapshots along the way, from the perspective of the user who is controlling the vehicle (such as the driver of a car), in order to avoid risk factors such as distraction and interference with driving control during the operation of the vehicle, the snapshot function of the camera on the vehicle cannot be applied to the snapshot control logic of general smart mobile terminals (such as mobile phones) and other devices. Therefore, in response to the snapshot needs of the above special scenarios, the industry currently adopts the following different solutions: using voice commands to trigger the camera shutter to capture, or storing the images collected along the way for the user to choose to capture later.

[0004] However, for the former, considering the speed of the vehicles in the scene, the user finds the target to be captured, then issues a voice command, which is then recognized and processed by the background, and finally triggers the shutter action of the shooting device. This process will take a relatively long time, resulting in the actual captured content no longer being the desired target; for the latter, in addition to the massive storage data taking up space and increasing the inconvenience of manual selection, the video streams recorded by the shooting device are mostly compressed, resulting in the imaging effect of the captured pictures often being unsatisfactory. Summary of the Invention

[0005] In order to solve one or more of the above-mentioned problems existing in the prior art, the present invention provides an image capturing method, apparatus, shooting equipment and vehicle.

[0006] The image capture method of the present invention includes: obtaining the user's shooting preferences based on historical photos and / or historical videos in the user's image collection; collecting images of the user's shooting device during its movement; and based on the user's shooting preferences, capturing a number of images from the images for storage.

[0007] According to an embodiment of the present invention, the capturing and storing of a plurality of images from the image includes: capturing and storing a plurality of images that at least meet the shooting preference from the image in combination with the prompt information.

[0008] According to an embodiment of the present invention, the prompt information further includes snapshot compliance information, and the prompt information is generated based on the shooting preference and user portrait.

[0009] According to an embodiment of the present invention, the prompt information further includes snapshot compliance information, which is used to constrain the operation of capturing images.

[0010] According to one embodiment of the present invention, the above-mentioned obtaining of the user's shooting preferences based on historical photos and / or historical videos in the user's image collection includes: obtaining the user's shooting preferences based on the picture features and / or shooting behavior features of the historical photos and / or historical videos.

[0011] According to one embodiment of the present invention, the image capture method further includes: pre-building a queue for temporarily storing captured images; storing the captured images in the queue in chronological order; and dynamically deleting the images in the queue when the number of images stored in the queue exceeds the queue length.

[0012] According to an embodiment of the present invention, the dynamic deletion of images in the queue includes: comparing the quality of all images in the queue, deleting the image with the lowest quality, and adjusting the remaining images in the queue in time sequence.

[0013] According to an embodiment of the present invention, the snapshot method further includes: upon receiving a snapshot instruction from a user, extracting at least a number of stored images within a preset period before the snapshot instruction as target snapshot images.

[0014] According to an embodiment of the present invention, the range of the preset time period is dynamically adjusted according to the real-time traveling speed.

[0015] The present invention also provides an image capture device, including: a capture preference analysis module, used to obtain the user's shooting preferences based on historical photos and / or historical videos in the user's image collection; a video stream acquisition module, used to capture images during the movement of the user's shooting device; and a candidate image capture module, used to capture a number of images from the images for storage based on the user's shooting preferences.

[0016] The present invention also provides a photographing device, comprising: a processing device, wherein the processing device is used to run stored computer-readable instructions, so that the photographing device executes the image capture method as described above.

[0017] The present invention also provides a vehicle, wherein the vehicle is equipped with at least one of the above-mentioned photographing devices.

[0018] As will be described in detail below, the present invention processes images captured by the user's camera during travel in real time in combination with the user's personalized shooting preferences, wherein the preference information is derived from at least the user's historical photos and / or historical videos. A number of images that meet the user's capture expectations are then actively captured from the real-time captured images and stored as alternatives. Subsequently, the user's instructions for triggering the capture can be monitored to extract a number of images as target capture images with original clarity. Alternatively, the user can select the target capture image from all images. The present invention effectively solves the problem of missing the capture object due to travel and response delays, and because the data size can be greatly reduced through the aforementioned capture link, it can fit into the existing storage space without the need for compression processing, thereby ensuring the clarity of the target capture image actually retained. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other features of the present invention will be described in detail below with reference to certain exemplary embodiments shown in the accompanying drawings, which are given below by way of illustration only and therefore do not limit the present invention, in which:

[0020] Figure 1 FIG. 1 is a schematic diagram showing the main process of an image capture method according to an embodiment of the present invention.

[0021] Figure 2 is a schematic flow chart illustrating a method for capturing candidate images according to an embodiment of the present invention;

[0022] Figure 3 is a schematic flow chart illustrating a method for storing candidate images according to an embodiment of the present invention;

[0023] Figure 4 FIG. 1 is a functional block diagram illustrating an image capture device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The present invention is described in detail below through specific embodiments so that those skilled in the art can easily implement the present invention according to the contents disclosed in this specification. The embodiments described below are only some embodiments of the present invention, not all. Based on the embodiments described in this specification, all other embodiments obtained by those skilled in the art without paying creative work are within the scope of protection of the present invention. It should be noted that the embodiments in this specification and the features in the embodiments can be combined with each other unless there is a conflict.

[0025] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "side", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention.

[0026] Furthermore, in the description of this specification, the order in which actions, processes, steps, and stages within a method are executed may be implemented in any order, unless otherwise expressly stated, such as "before," "before," or the results of a previous process are not used in a subsequent process. Even if the flow of actions in the claims, specification, and drawings is described using phrases such as "first" or "next" for convenience, this does not necessarily mean that they must be implemented in that order.

[0027] like Figure 1 Schematically, the image capture method of the present invention generally includes three steps: Step S0: obtaining the user's shooting preferences based on historical photos and / or historical videos in the user's image collection; Step S1: capturing images of the user's camera while the camera is moving; and Step S2: capturing and storing a number of images from the image collection based on the user's shooting preferences. The following describes these three steps in detail with reference to the accompanying figures.

[0028] Step S0: Acquire the user's shooting preference based on historical photos and / or historical videos in the user's image collection.

[0029] The user's shooting preference in the present invention is also the user's personalized information for the snapshot function. For example, the user's image collection (such as an album containing photos / videos) can be obtained in advance and the user's shooting preferences can be analyzed and predicted based on the historical photos and / or historical videos in the image collection (which can refer to image information that the user has taken or collected, etc.) to assist in forming the user's shooting preferences. Four points can be pointed out in this regard. First, the concept of the image collection is relatively broad and the source is not limited. For example, but not limited to the album content in the user's personal smart terminal, the album content in the network disk, the images in the storage medium, the images uploaded in real time, etc., provided by the user. However, considering personal privacy, the materials for analyzing the user's historical photos and videos must be actively selected and provided by the user or only allowed to be obtained from sources authorized by the user; second, the implementation method of using historical pictures and videos to analyze the user's shooting preferences can be considered in combination with algorithm models in the field of artificial intelligence (such as multimodal graph-to-text models), to use algorithms to analyze the user's existing images, infer their shooting behavior preferences and form user shooting preference information described in natural language. For example, after analyzing the user's historical photos, The analysis of the video obtained the following text to assist in constructing the user's shooting preference: "I like to use multiple photos to reflect travel experiences, feelings and discoveries, including shooting details, local customs, etc., to show a complete and rich travel record"; thirdly, when analyzing the user's historical photos and / or videos, it can be based on the picture features and / or shooting behavior features of these image materials, picture features such as composition, content, proportion, color tone, etc., shooting behavior features such as shooting time, shooting location, shooting frequency, number of similar photos, video shooting time, etc.; fourthly, it is understandable that the source of the user's shooting preference, in addition to the above-mentioned image collection, can also adopt preset instructions, real-time input, etc., which will be explained later.

[0030] Step S1: Capture images of the user's camera during its movement.

[0031] It should be pointed out that the technical concept of the present invention is not only reflected in its application to transportation vehicles. For example, in some scenarios, users can wear the shooting equipment on their heads, bodies, etc., and move in a variety of possible ways. When they need to take pictures while moving, they can also adopt the solution provided by the present invention.

[0032] In order to form a unified logical connection between the preceding and following texts, the following description will be specifically elaborated using the application in transportation as an exemplary scenario. However, it can be understood that transportation is not actually a limitation to the technical implementation of the present invention. Those skilled in the art can refer to the following description to reasonably associate different usage scenarios of mobile capture.

[0033] Therefore, the implementation of this step may not be limited to the location, number and type of the camera equipment. For example, it can be achieved by using cameras deployed in front, on the side and at the rear of the vehicle to capture video streams in different directions respectively, or it can be achieved by using a panoramic shooting device to capture a 360° video stream. The present invention does not limit this.

[0034] Step S2: Based on the user's shooting preference, capture a number of images from the image and store them.

[0035] This step can be specifically implemented by automatically and proactively capturing key frame raw images (uncompressed) relevant to the user's snapshot requirements from the captured video stream while the vehicle is in motion, based on the user's need for the snapshot function. This provides an alternative basis for the intelligent snapshot function. This embodiment does not limit the further processing method after obtaining the above-mentioned images, because these images can already be considered "snapshot images" that meet the user's shooting preferences and other personalized snapshot requirements. Therefore, after obtaining these images, these stored images can be directly used as target snapshot images. However, from the user's perspective, further selection from the images automatically stored by the machine may better meet the user's actual needs. Therefore, a trigger snapshot command issued by the user during driving can be used to further extract the target snapshot image from the candidate images in response to the command (a more specific implementation scheme for this is provided later and is not further described here). Of course, the user can also select the desired image from the captured images (this is not limited or detailed in the present invention). Based on the above, in the embodiments of the present invention below, the images obtained in the capture step are collectively referred to as candidate images (this does not exclude the possibility of directly using the candidate images as the target snapshot image). Those skilled in the art can easily understand that the candidate images herein refer to an optional range in which the final target captured image is located. Therefore, this can be further expanded into the following two aspects for detailed description:

[0036] First, intelligently capture keyframes:

[0037] The processing unit of the video stream is a frame of images. Artificial intelligence technology (such as machine learning, etc.) can be used to analyze the features of each frame in the image in combination with prompt information and filter out candidate images. For example, after being constrained by prompt information, each frame of the image can be classified into candidate images that meet the established requirements and "discarded" images that do not meet the requirements. Therefore, in some preferred embodiments of the present invention, the method of grabbing several frames of original images from the image as candidate images can refer to Figure 2 Indicative, specifically including:

[0038] Step S21: inputting each frame of original image in the image into the trained first model;

[0039] In actual operation, the first model can be but is not limited to a multimodal image-text question-answering model. For the multimodal image-text question-answering model, there are mature pre-trained large models in this field as precedents for image processing. Therefore, using this model can greatly reduce the model construction cost required by the present invention. The first model mentioned here is to distinguish it from the multimodal image-text model mentioned above. Accordingly, the model mentioned above involving the analysis of the user's historical image set can be called the second model in the present invention.

[0040] Step S22: The first model combines the given prompt information to filter and output the candidate image from each frame of the original image, wherein the prompt information refers to the user personalized information used for snapshot, that is, the personalized snapshot information mentioned above.

[0041] As mentioned above, the first model is preferably a multimodal image-text question-answering model. Therefore, in addition to the original images of each frame of the image as input, text-level information (i.e., the aforementioned prompt information) can also be used as input for model processing. This technical idea can refer to the processing process of the current large language model. In order to obtain processing results that are more in line with user needs, it is usually possible to give a number of limited prompt conditions (Prompt) to guide the model to generate more accurate and targeted outputs. That is, in this embodiment, the behavior of the first model in processing each frame of the image is controlled by text information input, so that it is more adapted to the needs of this application scenario.

[0042] Based on this concept, the present invention proposes in other embodiments that the prompt information can be derived from multi-dimensional user personalized information related to the snapshot requirements. For example, the prompt information can be generated using user shooting preferences described based on natural language and / or preset user portraits. The user shooting preferences described here can be the results of the pre-analysis and textual processing of the user's photo album mentioned above, but in actual operation it is not limited to this. For example, it can also be input by the user. To elaborate on this: Regarding the user shooting preferences, the user can, but is not limited to, actively set what shooting content he is interested in and the composition requirements in an offline manner. In actual operation, the user is allowed to flexibly set the shooting requirements in natural language modes such as voice input and text entry to form the user shooting preferences. For example, the user pre-enters the following transcribed text in voice mode: "I like magnificent scenery, and can create a dramatic atmosphere through light. I can also incorporate humanistic elements, but avoid modern content, such as passing vehicles." Regarding the user portrait, the user can also set his age, gender, occupation, ethnicity, language, interests and other multi-dimensional information offline. For example, the user pre-enters the following text "Gender: Male, Age: 35, Occupation: Athlete, Hobbies: Raising Animals and Plants" in a soft keyboard.

[0043] Based on the protection of personal privacy mentioned above, it is further taken into consideration that the process of actively capturing candidate images using a machine should comply with relevant laws and regulations. Therefore, in some better embodiments, the present invention proposes that the prompt information also includes pre-given snapshot compliance information as a constraint condition. For example, shooting taboos can be preset according to local requirements when the shooting equipment leaves the factory and used as one of the elements of the aforementioned prompt information (Prompt). For example, it can include privacy constraints such as the clear face of the photographed person, and confidential constraints such as specific areas or buildings to meet product compliance requirements. Of course, in actual operation, the above-mentioned snapshot compliance information can also be provided in text form in the form of natural language description. The present invention will not elaborate on this, but it can be further pointed out that for several elements that generate the prompt information, user shooting preferences, user portraits, etc., can be dynamically adjusted in actual applications, that is, users are allowed to change such prompt information as needed, and for compliance information, it is preferably solidified content that cannot be changed without authorization by unauthorized users.

[0044] Based on the above, it can be understood that when the background captures key frames from the video stream, the first model can filter out candidate images that meet the user's shooting preferences, user portraits and other personalized factors and meet the compliance constraints from each frame of input image under the above-mentioned multi-dimensional "prompt information" framework. In addition, it can be added that based on the development of artificial intelligence technology, the first model can also be interactively determined in advance in its task attributes. For example, before setting the above-mentioned prompt information, the following information can be provided to the first model: "You are an image analyzer. You can analyze whether each image meets my needs based on the given prompt information and classify it into high compliance, medium compliance, and low compliance. Images with medium compliance or above are automatically reserved. My prompt information includes..." In this way, the first model understands its task objectives and can analyze and select each frame of input image in combination with the prompt information as described above, so as to make the processing for the scenario of the present invention more intelligent.

[0045] Second, temporarily store candidate images:

[0046] In actual operation, the number of candidate images captured may not be limited. That is, if there are only a few video frames that meet the user's personalized information along the way of the vehicle, only a small number of candidate images may be captured. Conversely, a large number of candidate images may also be captured. However, considering the actual engineering implementation, shooting equipment performance, system cost and subsequent user capture operations, the number of candidate images captured may be limited. Figure 3 For example, in some preferred embodiments of the present invention, a method for temporarily storing candidate images is proposed, including:

[0047] Step S210: pre-building a queue for temporarily storing the candidate images;

[0048] Step S220: storing the captured candidate images into the queue in chronological order;

[0049] Step S230: When the number of candidate images stored in the queue exceeds the queue length, the candidate images in the queue are dynamically deleted according to a predetermined rule.

[0050] To elaborate, a queue for temporarily storing the candidate images can be pre-constructed, and the continuously captured candidate images can be stored in the queue in chronological order. Those skilled in the art will understand that this queue has a certain length limit (which can be set as needed). Therefore, when the number of candidate images that are continuously and automatically captured and stored in the queue during driving exceeds the queue length threshold, the candidate images in the queue can be dynamically deleted according to established rules (similar to the first model mentioned above that discards the screened-out non-candidate images when processing a single-frame image), providing space for candidate images that are subsequently added to the queue. The established rules mentioned here can be used in a conventional manner, such as first deleting the candidate image that was stored in the queue the earliest. Of course, it can also be combined with an intelligent algorithm to compare the quality of all the candidate images in the queue, delete the candidate image with the lowest quality, and fill the queue vacancies in time (that is, if a candidate image with relatively low quality is deleted from the middle of the queue, the vacancy caused by this operation is shifted and filled by the subsequent candidate image already in the queue). The quality mentioned here can preferentially or exclusively refer to the degree of compliance with the personalized information related to the user's snapshot requirements. In practical applications, the output of the first model mentioned above can be used to configure a corresponding score (such as a correlation probability value) for each selected candidate image. Therefore, when dynamically deleting images from the queue at this stage, the quantitative index of each candidate image in the queue can be directly compared to determine the candidate image with the relatively lowest quality. Optionally, the judgment of image quality can also take image clarity as a factor of the above-mentioned quantitative index based on the above-mentioned degree of compliance, thereby obtaining a comprehensive score of image quality, and then dynamically deleting images according to the comprehensive score. It can be further explained here that in order to ensure that candidate images continuously collected from the video stream are not missed, the first model combined with the processing mechanism of prompt information can be appropriately "relaxed". For example, the requirements can be relatively lowered when setting the aforementioned quality score threshold, so as to cover a wider range of candidate images.

[0051] Continuing from the previous article, after obtaining the uncompressed candidate images, it is possible to consider further reducing the range of the target snapshot images while driving. Therefore, in other preferred embodiments of the present invention, when a snapshot command issued by the user is monitored during the driving process, at least several of the candidate images within a preset time period before the snapshot command are extracted as target snapshot images.

[0052] Regarding the snapshot instruction, in actual application, it can refer to but is not limited to user voice instructions, body instructions (gestures, eye movements, etc.), and soft / hard switch signals triggered by the user. Of course, the present invention does not exclude other pre-set automatic snapshot trigger instructions.

[0053] In terms of the voice commands issued by the user, as mentioned above, combined with the scene of the vehicle driving, when the user discovers the object to be captured through observation and issues a voice command to start the capture function, then after the voice is received, recognized, and the intention is understood, the shooting device is triggered to capture, and the captured image actually obtained may miss the capture object that the user really wants to capture. Therefore, the present invention proposed in the above text to automatically collect candidate images that meet the needs of the user, and has intelligently included a number of images with time sequence information that meet the user's shooting preferences. Then, as long as the capture command actually given by the user is monitored, the candidate images within a predetermined time period before the capture command (which can refer to the moment t before the background of the shooting system receives the capture command) can be extracted from the candidate image range. The extracted candidate images can be additionally stored as the target capture images that are finally provided to the user, and can be set not to be deleted before the user actually views and selects the target capture image in the later stage. The above process can be specifically supplemented by the following three points:

[0054] First, the concept of the preset time period is relatively broad and is not limited to a time span range, but can also have a single time node. In addition, it can be pre-calibrated and pre-configured in the system background parameter list of the shooting device, or better yet, it can also be dynamically planned according to the real-time speed of the vehicle during travel, that is, the interval range of the preset time period can be adjusted according to the real-time speed of the vehicle; the advantage of this preferred example is that the real vehicle driving scene may produce different travel speeds due to various factors, and the capture scene that the present invention focuses on is strongly related to the travel speed. Dynamic adjustment of the aforementioned preset time period can, to a certain extent, avoid missing the capture object or failing to extract valid candidate images due to speed differences.

[0055] Secondly, in actual operation, this embodiment does not limit the number of images captured within the preset time period, nor does it limit whether it is necessary to collect captured images after receiving the capture instruction. For example, in other embodiments, in addition to extracting candidate images within the aforementioned preset time period, candidate images within a period of time after the capture instruction can also be continuously extracted; the advantage of this preferred example is that, since the shooting of the video stream is a continuous and coherent behavior, and in some scenarios, the capture object desired by the user may also be continuous content, therefore, based on the candidate images extracted before the user's capture instruction, each frame of the image continues to be captured as a candidate image and the target capture image is continuously extracted therefrom, which can provide more capture materials for subsequent users to view and select by themselves. Thus, neither omissions nor unexpected target images can be provided, which can greatly improve the user experience.

[0056] Third, in addition to the instructions issued by the user that need to be recognized and processed (voice, gestures, etc.), for the switch signal instructions triggered by the user (such as pressing the switch button pre-configured on the steering wheel or handle), similar to the above description, even if the switch signal does not need to be recognized and processed, the present invention still emphasizes that at least the candidate image before the electrical signal instruction must be extracted. This also takes into account the delay effect between human eye observation and the issuance of the switch instruction. Therefore, extracting several images before the instruction can ensure that the captured object actually seen by the user is not missed.

[0057] In summary, the technical solution according to the present invention can specifically have the following technical effects: the present invention processes the images captured by the user's camera during travel in real time in combination with the user's personalized snapshot preferences, wherein the preference information is at least derived from the user's historical photos and / or historical videos, and then actively captures a number of images that meet the user's snapshot expectations from the real-time captured images for storage as alternatives. Subsequently, by monitoring the user's instructions to trigger the snapshot, a number of images can be extracted as target snapshot images with original clarity, or the user can select the target snapshot image from all images. The present invention effectively solves the problem of missing the snapshot object due to travel and response delays, and because the data size can be greatly reduced through the aforementioned capture link, it can adapt to the existing storage space without performing compression processing, thereby ensuring the clarity of the target snapshot image actually retained.

[0058] Hereinafter, an image capturing device for implementing the above method will be further described. Figure 4 is a functional block diagram illustrating an image capture device according to an embodiment of the present invention, wherein the image capture device 400 includes:

[0059] The snapshot preference analysis module 401 is used to obtain the user's shooting preference based on historical photos and / or historical videos in the user's image collection;

[0060] The video stream acquisition module 402 is used to capture images of the user's camera during its movement;

[0061] The candidate image capturing module 403 is configured to capture and store a number of images from the image based on the user's shooting preference.

[0062] In a possible implementation of the above device embodiment, the candidate image capture module is configured to: capture and save a number of images that at least meet the shooting preference from the image in combination with the prompt information.

[0063] In a possible implementation of the above device embodiment, the prompt information is generated by the shooting preference and the user portrait.

[0064] In a possible implementation of the above device embodiment, the prompt information further includes pre-given snapshot compliance information for constraining an operation of capturing an image.

[0065] In a possible implementation of the above device embodiment, the snapshot preference analysis module is configured to obtain the user's shooting preference based on picture features and / or shooting behavior features of historical photos and / or historical videos.

[0066] In a possible implementation of the above device embodiment, the capture device further includes:

[0067] A queue building unit, used to pre-build a queue for temporarily storing captured images;

[0068] a candidate temporary storage unit, configured to store captured images into the queue in time sequence;

[0069] The queue adjustment unit is used to dynamically delete the images in the queue according to a predetermined rule when the number of images stored in the queue exceeds the queue length.

[0070] In a possible implementation of the above device embodiment, the queue adjustment unit is configured to: compare the quality of all the images in the queue, delete the image with the lowest quality and adjust the remaining images in the queue in time sequence.

[0071] In a possible implementation of the above-mentioned device embodiment, the capture device also includes: a target image extraction module, which is used to extract at least several stored images within a preset time period before the capture instruction as target capture images when receiving the capture instruction issued by the user.

[0072] In a possible implementation of the above device embodiment, the target image extraction module includes an extraction period setting unit, which is configured to dynamically adjust the interval range of the preset period according to the real-time travel speed.

[0073] Based on the description of the aforementioned method and device embodiments and the introduction of their effects, the present invention also provides a shooting device, which includes: a processing device, which is used to run stored computer-readable instructions so that the shooting device executes the image capture method described above.

[0074] To elaborate on this, the processing device is the hardware embodiment of the system background mentioned above. As the computing core of the shooting device, it can analyze data, instructions, etc., and can also serve as the control core of the shooting device. Therefore, the shooting device can refer to but is not limited to any component that can take photos and record videos, such as smart mobile terminals, photographic equipment, visual imaging components, etc., which have the aforementioned processing device.

[0075] In addition, the present invention further provides a vehicle, namely the vehicle mentioned above, the main feature of which is that it is equipped with at least one of the above-mentioned photographing devices, which needs to be explained in the following two points:

[0076] First, the configuration mode of the camera device is that the camera device can be installed on the vehicle at the factory, or it can be installed on the vehicle by the user or a third party.

[0077] Second, the term "vehicle" can be implemented in a variety of ways, including land vehicles such as motor vehicles, non-motor vehicles, and rail vehicles; water vehicles such as ferries and yachts; and air vehicles such as airplanes and unmanned aerial vehicles. It should be noted that drones can carry objects and thus can be considered vehicles for the purposes of this invention. Of course, this invention does not preclude the possibility of considering them as a type of flying photography equipment, based on different definitions. Therefore, within the context of this invention, the term "vehicle" can also reasonably be expanded to include mobile equipment for transporting materials, etc.

[0078] It is understood that the structures shown in the drawings are for illustration only and may include more or fewer modules or components than shown in the drawings, or have configurations different from those shown in the drawings. It is worth noting that when implementing the present invention using embodiments not exhaustively listed in this specification, those skilled in the art may adaptively adjust the structure, position, or functional arrangement of the relevant components.

[0079] It should be understood that, where technically feasible, the technical features listed above for different embodiments may be combined with each other to form additional embodiments within the scope of the present invention. In addition, the specific examples and embodiments described herein are non-limiting, and the structures, dimensions, and materials described above may be modified accordingly without departing from the scope of protection of the present invention.

[0080] In the present invention, the use of disjunctive conjunctions is intended to include conjunctions. The use of definite or indefinite articles is not intended to indicate cardinality. Specifically, reference to "the" object or "a" and "an" objects is intended to indicate a possible one of a plurality of such objects. In addition, the conjunction "or" can be used to convey simultaneous features rather than mutually exclusive solutions. In other words, the conjunction "or" should be understood to include "and / or". The term "include" is inclusive and has the same scope as "comprise".

[0081] The above embodiments, particularly any "preferred" embodiments, are possible examples of implementations and are presented merely for a clear understanding of the principles of the present invention. Many changes and modifications may be made to the above embodiments without departing substantially from the spirit and principles of the technology described herein. All modifications are intended to be included within the scope of the present invention.

[0082] All documents mentioned in this specification are incorporated herein by reference, just as if each document were incorporated herein by reference in its entirety.

[0083] In addition, it should be understood that after reading the above description of the present invention, those skilled in the art may make various changes or modifications to the present invention, and these equivalent forms also fall within the scope of protection of the present invention.

Claims

1. An image capture method, characterized in that: include: Obtaining the user's shooting preferences based on historical photos and / or historical videos in the user's image collection; Collect images of the user's camera while it is moving; Based on the user's shooting preference, a number of images are captured from the image and stored.

2. The image capture method according to claim 1, wherein: The capturing of a plurality of images from the image for storage comprises: In combination with the prompt information, a plurality of images that at least meet the shooting preference are captured and saved from the image.

3. The image capture method according to claim 2, wherein: The prompt information is generated based on the shooting preference and user portrait.

4. The image capture method according to claim 3, wherein: The prompt information also includes snapshot compliance information, which is used to restrict the operation of capturing images.

5. The image capture method according to claim 1, wherein: The acquiring of the user's shooting preferences based on historical photos and / or historical videos in the user's image collection includes: The user's shooting preference is obtained based on the picture features and / or shooting behavior features of the historical photos and / or historical videos.

6. The image capture method according to claim 1, wherein: Also includes: Pre-build a queue for temporarily storing captured images; Storing the captured images into the queue in time sequence; When the number of images stored in the queue exceeds the queue length, the images in the queue are dynamically deleted.

7. The image capture method according to claim 6, wherein: The dynamic deletion of images in the queue includes: comparing the quality of all images in the queue, deleting the image with the lowest quality, and adjusting the remaining images in the queue in time sequence.

8. The image capture method according to any one of claims 1 to 7, characterized in that: The snapshot method further includes: upon receiving a snapshot instruction issued by a user, extracting at least a number of stored images within a preset time period before the snapshot instruction as target snapshot images.

9. The image capture method according to claim 8, wherein: The interval range of the preset time period is dynamically adjusted according to the real-time travel speed.

10. An image capture device, characterized in that: include: A snapshot preference analysis module is used to obtain the user's shooting preferences based on historical photos and / or historical videos in the user's image collection; The video stream acquisition module is used to capture images of the user's camera during its movement; The candidate image capture module is used to capture and store a number of images from the image based on the user's shooting preference.

11. A photographing device, characterized in that: The apparatus comprises a processing device configured to execute stored computer-readable instructions so as to enable the photographing device to execute the image capturing method according to any one of claims 1 to 9.

12. A means of transport, characterized in that: At least one photographing device according to claim 11 is arranged on the vehicle.

Images