WO2022087826A1

WO2022087826A1 - Video processing method and apparatus, mobile device, and readable storage medium

Info

Publication number: WO2022087826A1
Application number: PCT/CN2020/123998
Authority: WO
Inventors: 周娴; 李鑫超; 刘志鹏
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2020-10-27
Filing date: 2020-10-27
Publication date: 2022-05-05
Also published as: CN114026874A

Abstract

A video processing method and apparatus, a mobile device, and a readable storage medium. The method comprises: for a target picture in a video to be processed, determining target information corresponding to a picture parameter of the target picture, the target information being used for representing whether the picture parameter of the target picture satisfies a preset requirement (101); according to the target information of the target picture comprised in the video to be processed, performing screening processing on the video to be processed, wherein the screening processing comprises deleting the target picture that does not satisfy the preset requirement or one of video clips to which the target picture belongs (102); and finally, according to the video to be processed subjected to the screening processing, generating a target video (103). By means of the present invention, during video processing, pictures in a video are automatically screened according to picture parameters of the pictures, so that the screening cost can be reduced to a certain extent, and the screening efficiency is improved.

Description

Video processing method, apparatus, removable device and readable storage medium

technical field

The present invention belongs to the field of network technologies, and in particular, relates to a video processing method, device, removable device and readable storage medium.

Background technique

At present, video is an excellent way to obtain information, and video is produced in more and more scenarios. In order to improve the quality of the video, it is often necessary to filter the pictures with poor quality in the video. In the existing methods, screening is often performed directly through manual screening. However, this screening method is costly and inefficient.

SUMMARY OF THE INVENTION

The present invention provides a video processing method, device, removable device and readable storage medium, so as to solve the problems of high screening cost and low screening efficiency.

In order to solve the above-mentioned technical problems, the present invention is achieved in this way:

In a first aspect, an embodiment of the present invention provides a video processing method, which includes:

For the target picture in the video to be processed, determine the target information corresponding to the picture parameter of the target picture; the target information is used to represent whether the picture parameter of the target picture meets the preset requirements;

Screening the video to be processed according to the target information of the target picture contained in the video to be processed; wherein, the screening process includes deleting target pictures that do not meet the preset requirements or the target pictures to which the target pictures belong. one of the video clips;

Generate a target video according to the to-be-processed video after screening.

In a second aspect, an embodiment of the present invention provides a video processing apparatus, and the apparatus includes a memory and a processor;

the memory for storing program codes;

The processor calls the program code, and when the program code is executed, is configured to perform the following operations:

Generate a target video according to the to-be-processed video after screening.

In a third aspect, an embodiment of the present invention provides a movable device, where the movable device is configured to execute the steps in the video processing method described in the first aspect.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the following operations are implemented:

Generate a target video according to the to-be-processed video after screening.

In the embodiment of the present invention, for the target picture in the video to be processed, target information corresponding to the picture parameters of the target picture is determined, wherein the target information is used to indicate whether the picture parameters of the target picture meet the preset requirements. The target information of the included target picture is screened for the video to be processed, wherein the screening process includes deleting the target picture that does not meet the preset requirements or one of the video clips to which the target picture belongs, and finally, according to the screened and processed video to be processed , generate the target video. In this way, by automatically screening the pictures in the video according to the picture parameters of the pictures during the video processing, the screening cost can be reduced to a certain extent and the screening efficiency can be improved.

Description of drawings

1 is a flowchart of steps of a video processing method provided by an embodiment of the present invention;

2 is a schematic diagram of a clipping process provided by an embodiment of the present invention;

3 is a block diagram of a video processing apparatus provided by an embodiment of the present invention;

FIG. 4 is a block diagram of a computing processing device according to an embodiment of the present invention;

FIG. 5 is a block diagram of a portable or fixed storage unit according to an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

FIG. 1 is a flowchart of steps of a video processing method provided by an embodiment of the present invention. As shown in FIG. 1 , the method may include:

Step 101: For a target picture in the video to be processed, determine target information corresponding to a picture parameter of the target picture; the target information is used to represent whether the picture parameter of the target picture meets a preset requirement.

In this embodiment of the present invention, the video to be processed may be a video that needs to be screened for pictures with poor quality. For example, the video to be processed may be a video shot by the user, or a video downloaded from a network, which is not limited in this embodiment of the present invention. Further, a video can be essentially understood as a picture sequence composed of multiple frames of images, and the target picture in the video to be processed may be all pictures in the picture sequence, or a part of the pictures in the picture sequence. The picture parameter of the target picture may be a parameter that can characterize the picture quality, and the specific number and type of the picture parameter may be set according to actual requirements. For example, the picture parameters may be set as the clarity of the picture, the exposure level of the picture, and the like.

Further, the preset requirement may be a quality requirement that needs to be satisfied by the picture parameters of the picture. If the picture parameters of the picture meet the preset requirements, it means that the picture quality is high and can meet the quality requirements. On the contrary, if the picture parameters of the picture do not meet the preset requirements, it means that the quality of the picture is poor and cannot meet the quality requirements. The specific content of the preset requirements can be set according to actual needs. For example, the preset requirement may be that the degree of sharpness is greater than the preset sharpness threshold, or the degree of exposure is within the range of the preset exposure degree, and so on. Further, the target information may be a tag that can characterize whether the picture parameters of the target picture meet the preset requirements. The specific content of the label can be set according to actual needs. For example, numbers, letters, special symbols, etc. can be used as labels. The specific content of the label when the picture parameters meet the preset requirements is different from the specific content of the label when the picture parameters do not meet the preset requirements. For example, "0" may be used as a label indicating that the picture parameters of the target picture do not meet the preset requirements, and "1" may be used as a label indicating that the picture parameters of the target picture meet the preset requirements.

Step 102: Perform screening processing on the video to be processed according to the target information of the target picture contained in the video to be processed; wherein, the screening processing includes deleting target pictures or the target pictures that do not meet the preset requirements One of the video clips the image belongs to.

When performing the screening process in this embodiment of the present invention, for each target picture, whether the picture parameters of the target picture satisfy a preset condition may be first determined according to the target information of the target picture. For example, it may be determined that the picture parameter does not meet the preset requirement when the target information is 0, and it can be determined that the picture parameter meets the preset requirement when the target information is 1. Further, if the picture parameters of the target picture do not meet the preset requirements, the target picture may be determined as the target picture that does not meet the preset requirements. Correspondingly, the target picture may be deleted or one of the video clips to which the target picture that does not meet the preset requirements belongs may be directly deleted. One of the video clips to which the target picture belongs may be a video clip that includes the target picture in the video to be processed, and the specific length of the video clip can be set according to actual requirements, which is not limited in this embodiment of the present invention.

Step 103 , generating a target video according to the to-be-processed video after screening and processing.

In the embodiment of the present invention, firstly, according to the target information of the target picture in the video to be processed, the video to be processed is screened, which can reduce the pictures of poor quality included in the video to be processed, and further improve the image of the video to be processed after the screening process. quality. Correspondingly, after the screening is completed, the target video is generated according to the video to be processed after screening, which can ensure that the final generated target video has higher image quality. For example, when the target video is generated, the screened video to be processed may be directly used as the target video.

To sum up, the video processing method provided by the embodiment of the present invention can determine the target information corresponding to the picture parameter of the target picture for the target picture in the video to be processed, wherein the target information is used to indicate whether the picture parameter of the target picture satisfies the According to the preset requirements, the to-be-processed video is screened according to the target information of the target picture contained in the to-be-processed video, wherein the screening process includes deleting the target picture that does not meet the preset requirements or one of the video clips to which the target picture belongs, and finally , and generate the target video according to the to-be-processed video after screening. In this way, by automatically screening the pictures in the video according to the picture parameters of the pictures during the video processing, the screening cost can be reduced to a certain extent and the screening efficiency can be improved.

Optionally, in an implementation manner of the embodiment of the present invention, the above picture parameters may include one or more of a blur degree, a shake degree, an exposure degree, and a color change degree of the picture. Certainly, the picture parameters may also include other types of parameters, which are not limited in this embodiment of the present invention.

Among them, based on the degree of blurring of the picture, it can be determined whether the overall picture is clear, based on the degree of shaking of the picture, it can be determined whether the picture of the picture has a shaking problem, based on the degree of exposure of the picture, it can be determined whether the brightness of the picture is appropriate, and the degree of color change of the picture can be determined. Is the color difference of the picture appropriate? However, in practical application scenarios, when the picture is blurred and unclear, the picture is shaken, the brightness is too high or too low, and the color difference is too high or too low, it often means that the quality of the picture is poor. Therefore, in the embodiment of the present invention, by setting the degree of blurring, the degree of shaking, the degree of exposure and/or the degree of color change of the picture as the picture parameters, the pictures in the video to be repaired can be screened from these parameters in the future, and then certain To a certain extent, it can ensure that pictures with poor quality can be deleted more accurately.

It should be noted that the more types of picture parameters, the more dimensions will be referenced in subsequent screening, and accordingly, the final screening results will be more accurate, but the amount of calculation will also increase accordingly. Conversely, if there are fewer types of picture parameters, the less dimensions will be referenced in subsequent screening, and accordingly, the accuracy of the final screening results will be reduced, but the amount of calculation will be smaller. Therefore, in the specific implementation, the calculation amount that the hardware device can bear may be determined first, and the number of selected picture parameters is determined according to the amount of calculation that the hardware device can bear, so as to ensure that the hardware device can run normally. Wherein, the number of selected picture parameters is positively related to the loadable calculation amount.

It should be noted that, in the case where multiple picture parameters are selected, there may be multiple pieces of target information corresponding to the picture parameters. As an example, assuming that the degree of blurring, the degree of shaking, the degree of exposure, and the degree of color change are selected, then the fuzzy label information corresponding to the degree of blur, the information of the shaking label corresponding to the degree of shaking, the information of the exposure label corresponding to the degree of exposure, and the corresponding degree of color change can be obtained. color difference label information. Among them, the fuzzy label information, the shake label information, the exposure label information and the color difference label information are the target information. The fuzzy label information can be used to characterize whether the blur degree of the target image meets the preset requirement corresponding to the blur degree, the jitter label information can be used to characterize whether the jitter degree of the target image meets the preset requirement corresponding to the jitter degree, and the exposure label information is used to characterize the target Whether the exposure degree of the picture meets the preset requirements corresponding to the exposure degree, and the color difference label information is used to indicate whether the color change degree of the target image meets the preset requirements corresponding to the color change degree.

Optionally, in this embodiment of the present invention, the above-mentioned operation of screening the video to be processed according to the target information of the target picture contained in the video to be processed may be implemented by the following sub-steps:

Sub-step (1): segment the to-be-processed video to obtain multiple video segments.

For example, the video to be processed may be divided into video segments containing the same number of frames by using a preset segmentation algorithm in a manner of dividing at equal intervals. Alternatively, a preset segmentation algorithm may be used to randomly divide the video to be processed into video segments containing different numbers of frames in a manner of unequal interval division, which is not limited in this embodiment of the present invention. Of course, other division manners may also be used to implement division, which is not limited in this embodiment of the present invention.

Sub-step (2): For each of the video clips, perform screening processing on the video clips according to the target information of the target picture contained in the video clips.

In this step, a video clip may be used as a processing unit, and screening processing may be performed for each video clip. The screening process for the video clip may include deleting the target picture or the video clip that does not meet the preset requirements in the video clip.

In the embodiment of the present invention, when the screening process is specifically performed, the video to be processed is first divided to obtain a plurality of video segments. Next, according to the target information of the target picture contained in the video clip, the video clip is used as a processing unit to filter the video clip. In this way, by dividing the video to be processed into video segments with smaller granularity and then processing, the burden of each processing operation can be reduced to a certain extent, and further processing efficiency can be ensured to a certain extent.

Further, in the case of dividing the video to be processed into video segments, in the subsequent operation of generating the target video according to the video to be processed after the screening process, the video clips after the screening process can be merged to obtain the target video. Since the screening process deletes pictures with poor quality in each video clip, the target video is obtained by merging the video clips after the screening process, which can ensure the quality of the target video to a certain extent. Specifically, when merging, the filtered video segments may be merged according to the sequence in the video to be processed, so as to ensure that the final target video can be played normally.

Optionally, in an embodiment of the present invention, the video to be processed may be independently selected by the user. Specifically, before the target information corresponding to the picture parameter of the target picture is determined, a user's selection operation on an optional video may be received; the optional video selected by the selection operation may be determined as the to-be-processed video; the optional video For videos stored in electronic devices. Correspondingly, after the target video is generated, the target video can be displayed to the user, so as to ensure that the user can obtain the processing result in time and improve the interaction effect.

Wherein, each video stored in the electronic device can be displayed in the client, and the user can click the optional video to be processed to realize the input selection operation. Correspondingly, the electronic device may receive the selection operation, and determine the selected optional video as the video to be processed. It should be noted that, in this embodiment of the present invention, after the target information is determined, target information may be added to each target picture and displayed to the user. The user can select the target image to be deleted according to the requirements and in combination with the target information of the target image. Accordingly, the electronic device can delete the target picture selected by the user. In this way, by displaying the target picture with the target information added to the user, the user can independently decide the specific target picture to be deleted, which can improve the flexibility of the operation.

Optionally, in this embodiment of the present invention, for the target picture in the video to be processed, before the operation of determining the target information corresponding to the picture parameter of the target picture, the following steps may also be performed:

Step A: Determine the target picture according to the pictures contained in the video to be processed; extract the target picture from the video to be processed, and add timestamp information to each of the target pictures; the timestamp information Used to characterize the order of the target picture in the target picture.

In this step, the appearance time point of each target picture in the video to be processed may be used as timestamp information. For example, it is assumed that the target picture includes picture a, picture b, and picture c. Among them, the appearance time points of the picture a, the picture b, and the picture c are respectively: the 10th second, the 15th second, and the 21st second. Then, the 10th second, the 15th second, and the 21st second can be used as the timestamp information including the picture a, the picture b, and the picture c, respectively. Further, when adding time stamp information to each target picture, an association relationship between each target picture and the corresponding time stamp information may be established, or, the corresponding time stamp information may also be written into the target picture. .

Correspondingly, after the operation of determining the target information corresponding to the picture parameter of the target picture for the target picture in the video to be processed, the following steps can also be performed: Step B: According to the timestamp information of the target picture, the The target picture is merged into the to-be-processed video. During specific implementation, the location of the target picture may be determined according to the timestamp information first, and then the target picture is inserted into the corresponding location. For example, picture a may be inserted into the video to be processed at the 10th second, picture b may be inserted into the video to be processed at the 15th second, and picture c may be inserted into the video to be processed at the 21st second.

It should be noted that, in this embodiment of the present invention, before incorporating the target image into the video to be processed, it is also possible to determine a third target image whose parameters do not meet the preset requirements, and then delete the third target image. Wherein, the preset requirement may include one or more of normal blur degree, normal jitter degree, normal exposure degree and normal color change degree. Specifically, for each picture parameter, it can be first determined whether the target information corresponding to the picture parameter is a label used to indicate that the picture parameter does not meet the preset requirements, and if so, it can be determined that none of the picture parameters meet the preset requirements. . Further, if all the picture parameters do not meet the preset requirements, it means that the quality of the picture is very poor, and therefore, the third target picture can be directly deleted before being merged. In this way, the burden of subsequent screening processing can be reduced, and unnecessary target pictures can be avoided from performing merging operations, thereby ensuring processing efficiency.

In the embodiment of the present invention, before the operation of determining the target information corresponding to the picture parameters of the target picture, the target picture is determined and the target picture is extracted. In this way, by extracting the target picture separately, it is possible to avoid interference from other pictures in the video to be processed when the target information corresponding to the picture parameters of the target picture is subsequently determined, thereby ensuring the processing effect to a certain extent. At the same time, after the target information is determined according to the extracted target picture, the target picture is combined into the video to be processed, so as to ensure that the subsequent steps can be performed normally. And by adding time stamp information to the target picture, the target picture can be conveniently combined into the video to be processed according to the time stamp information, thereby ensuring the processing effect.

Optionally, in an embodiment of the present invention, according to the pictures contained in the video to be processed, the operation of determining the target picture can be implemented by the following sub-steps:

Sub-step (3): when the total number of pictures contained in the video to be processed is greater than the first preset number threshold, select m frames of pictures from the video to be processed according to a preset frame rate as the target picture; the m is not greater than the first preset number threshold.

In this step, the first preset number threshold and the preset frame rate may be set according to actual requirements. Wherein, the first preset number threshold may be the number of pictures required to implement screening. For example, the first preset number threshold may be 50, 60, and so on. The preset frame rate can be 4 frames per second (fps). Specifically, the total number of pictures included in the video to be processed may be determined first, for example, the configuration information of the video to be processed may be read to obtain the total number. Next, it is determined whether the total number is greater than the first preset number threshold. If the total number is greater than the first preset number threshold, it can be considered that the total number of pictures is large, and if all pictures are directly used as target pictures, too much calculation amount will be introduced. Therefore, selection can be made according to the preset frame rate, and accordingly, these target pictures can be correspondingly extracted subsequently. That is, frame extraction can be performed according to a preset frame rate to obtain m target pictures, and the m target pictures form a picture sequence.

Sub-step (4): under the condition that the total number is not greater than the first preset number threshold, use all pictures included in the video to be processed as the target picture.

If the total number is not greater than the first preset number threshold, it means that if all the pictures are directly used as the target pictures, too much calculation amount will not be introduced. Therefore, all pictures can be directly used as target pictures without other processing. In this way, it is possible to avoid performing unnecessary other processing while realizing the determination of the target picture.

In this embodiment of the present invention, all pictures included in the video to be processed are used as target pictures only when the total number is not greater than the first preset number threshold, and when the total number is greater than the first preset number threshold , select only part of the image as the target image. In this way, it can be ensured that there are enough target pictures to a certain extent while avoiding too many target pictures, so as to ensure that more information can be provided for subsequent screening, and the screening effect can be ensured.

Optionally, the preset requirements in this embodiment of the present invention may include preset requirements corresponding to the degree of blur: normal degree of blur, preset requirements corresponding to the degree of jitter: normal degree of jitter and preset requirements corresponding to the degree of exposure: normal degree of exposure, The preset requirements corresponding to the degree of color change: the degree of color change is normal.

Further, the target information corresponding to the blur degree of the target picture can be determined through the following sub-steps:

Sub-step (5): determine the first ratio of the first number of pixels with the fuzzy confidence greater than the preset reliability threshold in the target picture to the total number of pixels of the target picture; according to the first ratio and the first ratio; A magnitude relationship between preset ratio thresholds determines blur label information; the blur label information is used to represent whether the blur degree of the target image is normal.

In this step, the fuzzy confidence level can be used to represent the probability that the pixel point is a fuzzy pixel point. Specifically, the target image may be input into a preset blur detection module, and the blur detection module may be implemented based on a semantic segmentation network composed of convolutional neural networks (Convolutional Neural Networks, CNN). In practical application scenarios, sample images with different degrees of blur can be obtained, and the neural network can be trained by using these sample images to generate a blur detection module. Further, in use, the blur detection module can extract the picture features of the input target picture, and then determine the blur confidence of each pixel in the target picture according to the extracted picture features.

Further, the preset reliability threshold and the first preset ratio threshold may be set according to actual requirements. For example, the preset reliability threshold may be 80%. The first preset ratio threshold may be 0.2. Further, if the fuzzy confidence of the pixel point is greater than the preset confidence threshold, it can be considered that the pixel point has a fuzzy problem, and the pixel point is a fuzzy pixel point. Correspondingly, in this step, the fuzzy confidence of each pixel can be compared with a preset confidence threshold to determine the first number. Then, the first number is divided by the total number of pixels of the target image to obtain a first ratio.

If the first ratio is greater than the first preset ratio threshold, it may be considered that the degree of blurring of the target image is relatively serious, and accordingly, the blurring label information of the target image may be set as the first blurring label. Wherein, the first fuzzy label indicates that the degree of blurring of the target image is abnormal. If the first ratio is not greater than the first preset ratio threshold, it can be considered that the blur degree of the target image is within an acceptable range, and accordingly, the blur tag information of the target image can be set as the second blur tag. Wherein, the second fuzzy label indicates that the degree of blurring of the target image is normal. For example, the first fuzzy label may be "unclear", and the second fuzzy label may be "clear".

In the embodiment of the present invention, by determining the proportion of the blurred pixels in the target picture, whether the target picture is blurred is determined according to the size of the proportion, and corresponding fuzzy label information is set. To a certain extent, it can ensure the accuracy of the judgment result of image blurring, and ensure the credibility of the set fuzzy label information.

It should be noted that, in actual application scenarios, sometimes the subject is mainly focused during the shooting process, and the background is intentionally blurred to highlight the subject. Therefore, in the embodiment of the present invention, the background area in the target image can also be determined, and then the coincidence ratio of the blurred area and the background area can be detected; wherein, the blurred area is an area composed of pixels whose blur confidence is greater than a preset confidence threshold. In the case where the first ratio is greater than the first preset ratio threshold and the overlap ratio is not greater than the preset overlap ratio threshold, the blur tag information of the target image is set as the first blur tag. The preset coincidence ratio threshold can be set according to actual needs. For example, the preset coincidence ratio threshold can be 90%. If the coincidence ratio is greater than the preset coincidence ratio threshold, it can be considered that the blurred area in the target image is a normal phenomenon. If the coincidence ratio is not greater than the preset coincidence ratio threshold, it can be considered that the blurred area in the target image is caused by abnormal factors.

In this embodiment of the present invention, by detecting whether the blurred area is a background area, only when the blurred area is not a background area (the overlap ratio is not greater than the preset overlap ratio threshold) and the first ratio is greater than the first preset ratio threshold The target image sets a first blur label representing an abnormal blur degree. In this way, in the case of blurred background caused by focusing, it can be avoided that the target picture is misjudged as a picture with an abnormal degree of blurring, and then inappropriate fuzzy label information is set for the target picture.

Further, the target information corresponding to the degree of shaking of the target picture can be determined by the following sub-steps:

Sub-step (6): take the target picture as the input of the first preset classification model, and determine the jitter label information according to the output category of the first preset classification model; the first preset classification model is used for Whether the degree of jitter is normal for image classification.

In this step, the shaking label information is used to represent whether the shaking degree of the target image is normal. The first preset classification model may be based on a CNN neural network, and the first preset classification model may be obtained by training sample pictures with different degrees of jitter (including normal and abnormal degrees of jitter), and the first preset classification model may be. During the training process, through deep learning to learn the ability to distinguish whether the jitter of the picture is normal or not. Specifically, after the target picture is input into the first preset classification model, the first preset classification model can extract the picture features of the input target picture, and then according to the extracted picture features, determine whether the degree of shaking of the target picture is normal, and output category. Correspondingly, if the output category of the first preset classification model is a category representing an abnormal degree of shaking, the shaking label information of the target image may be set as the first fuzzy label. If the output category of the first preset classification model is a category representing a normal degree of shaking, the shaking label information of the target image may be set as the second fuzzy label. For example, the first jitter tag may be "tremble", and the second jitter tag may be "untremble".

In this embodiment of the present invention, the jitter label information is determined by the output category of the first preset classification model. In this way, it is only necessary to input the target image into the first preset classification model to conveniently determine whether the target image jitters, which in turn can facilitate Set the jitter label information later to improve the setting efficiency.

Further, the target information corresponding to the exposure degree of the target picture can be determined by the following sub-steps:

Sub-step (7): take the target picture as the input of the second preset classification model, and determine the exposure label information according to the output category of the second preset classification model; the second preset classification model is used for Whether the exposure level is normal for image classification.

In this step, the exposure label information is used to represent whether the exposure degree of the target image is normal. The second preset classification model may be based on a CNN neural network, and the second preset classification model may be obtained by training sample pictures with different exposure levels (including normal exposure and abnormal exposure). The second preset classification model The ability to distinguish whether the exposure level of a picture is normal can be learned during the training process. Specifically, after the target picture is input into the second preset classification model, the second preset classification model can extract the picture features of the input target picture, and then judge whether the exposure degree of the target picture is normal according to the extracted picture features, and output category. Correspondingly, if the output category of the second preset classification model is a category representing an abnormal degree of exposure, the exposure label information of the target image may be set as the first exposure label. If the output category of the second preset classification model is a category representing a normal exposure degree, the exposure label information of the target image may be set as the second exposure label. For example, the first exposure tag may be "expose F", and the second exposure tag may be "expose R".

In the embodiment of the present invention, the exposure label information is determined by the output category of the second preset classification model. In this way, it is only necessary to input the target image into the second preset classification model to conveniently determine whether the target image has abnormal exposure ( Overexposure and overdarkness), which can facilitate subsequent setting of exposure label information and improve setting efficiency.

Further, the target information corresponding to the color change degree of the target picture can be determined by the following sub-steps:

Sub-step (8): determine the second ratio between the second number of pixels whose color value exceeds the preset color value range in the target picture and the total number of pixels; according to the second ratio and the second preset ratio The size relationship between the thresholds determines the color difference label information; the color difference label information is used to represent whether the color change degree of the target picture is normal.

In the embodiment of the present invention, the color value of the pixel point may be the color channel value of the pixel point, for example, the red, green and blue (RGB) color channel value. The preset color value range and the second preset ratio threshold can be set according to actual needs. For example, the preset color value range may be determined according to the lowest color value and the highest color value in multiple pictures with normal color difference. The second preset ratio threshold may be based on the lowest ratio threshold that will affect the user's viewing experience. If the color value falls within the preset color value range, it can be considered that the color of the pixel point is normal, and if the color value does not fall within the preset color value range, it can be considered that the color of the pixel point is abnormal.

Further, a color value detection algorithm may be used to determine the color value of each pixel in the target image. The color value is then compared to a preset range of color values to determine the second quantity. The second ratio is then divided by the total number of pixels in the target image.

If the second ratio is greater than the second preset ratio threshold, it can be considered that the color difference of the target picture is normal, that is, the degree of color change is abnormal. Correspondingly, the color difference label information of the target picture may be set as the first color difference label. The first color difference label indicates that the color change degree of the target picture is abnormal. If the second ratio is not greater than the second preset ratio threshold, it can be considered that the color change degree of the target picture is normal, and accordingly, the color difference label information of the target picture can be set as the second color difference label. The second color difference label indicates that the color change degree of the target picture is normal. For example, the first color difference label may be "Color F", and the second color difference label may be "Color R".

In the embodiment of the present invention, by determining the proportion of pixels with abnormal colors in the target picture, determining whether the color change degree of the target picture is normal according to the proportion of the proportion, and setting the corresponding color difference label information, to a certain extent, it is possible to ensure that the picture quality The accuracy of the color difference judgment result, and the reliability of the set color difference label information.

Optionally, in the embodiment of the present invention, the above-mentioned step of screening the video clips according to the target information of the target pictures included in the video clips may include:

Sub-step (2a): Determine a third number of first target pictures in the video segment; the first target pictures are target pictures whose target information indicates that picture parameters do not meet the preset requirements and appear continuously.

In this step, the target pictures that appear continuously refer to the pictures that are adjacent to the target picture in the forward direction and/or the backward direction in the video segment are also target pictures. Specifically, the target pictures whose picture parameters do not meet the preset requirements may be determined according to the target information of each target picture in the video clip. Then, the pictures of the target pictures whose picture parameters do not meet the preset requirements that appear continuously are used as the first target pictures. Wherein, the picture parameters do not meet the preset requirements may be that all the picture parameters do not meet the preset requirements, or some picture parameters do not meet the preset requirements. Correspondingly, the target picture that all picture parameters do not meet the preset requirements and appears continuously can be used as the first target picture, for example, the target information can be: the first fuzzy label "unclear", the first jitter label "tremble", The consecutively appearing target pictures of the first exposure label "expose F" and the first color difference label "Color F" are used as the first target picture. Alternatively, a target picture in which some of the picture parameters do not meet the preset requirements and which appear continuously may also be used as the first target picture.

For example, it is assumed that the video clip includes picture 1 to picture 20. The target pictures whose picture parameters do not meet the preset requirements include: picture 1, picture 10, picture 11, and picture 13. Because the picture 1 to the picture 10 are the target pictures whose parameters that appear continuously do not meet the preset requirements. Therefore, picture 1 to picture 10 can be determined as the first target picture.

Sub-step (2b): if the third number is greater than the second preset number threshold, remove n of the first target pictures; the n is less than the third number.

In this step, the second preset number threshold may be set according to actual needs. If the third number is greater than the second preset number threshold, it can be considered that if all the first target pictures are directly deleted, there is a high probability that the video will appear in the video later. The front and back of the screen are not harmoniously connected, and the video playback is not smooth. Therefore, only part of the first target picture can be culled. Specifically, n first target pictures may be randomly selected for deletion.

In the embodiment of the present invention, by determining the first target pictures that appear continuously and the picture parameters do not meet the preset requirements, and in the case of a large number of first target pictures, n partial first target pictures are eliminated, and the continuous low-quality pictures can be In the case of occurrence, avoid the problem that the subsequent video playback is not smooth due to culling too many consecutive pictures.

Further, in this embodiment of the present invention, after removing n first target pictures, a picture transition effect may be added to the remaining first target pictures, wherein the picture transition effect may be used to control the first target picture The display effect when it appears, the picture transition effect can include one or more of right out, left in, spin out and spin in, and fade in and fade out. Certainly, the picture transition effect may also include other types of effects, which are not limited in this embodiment of the present invention. In this way, by adding a picture transition effect to the remaining first target picture, the display effect when the first target picture is displayed can be improved to a certain extent. It should be noted that, in this embodiment of the present invention, an operation of adding a picture transition effect may be performed when an adding instruction sent by a user is received. In this way, unnecessary adding operations can be avoided, resulting in that the final generated target video cannot satisfy the user’s needs. question of needs.

Further, in this embodiment of the present invention, after removing n first target pictures, a transition video frame may be generated according to the picture content of the remaining first target pictures; wherein, the picture parameters of the transition video frame meet preset requirements and The picture similarity is greater than the preset similarity threshold, and the picture similarity is the similarity between the picture content of the transition video frame and the picture content of the remaining first target picture. The preset similarity threshold may be set according to actual requirements. For example, the preset similarity threshold may be 99%. Next, transition video frames are added to the remaining first target pictures. In the embodiment of the present invention, by further inserting high-quality pictures with similar content into the remaining first target pictures, the image quality of the video can be ensured to a certain extent, and at the same time, the culling of the first target picture can be avoided to a greater extent. In the video, the front and back of the video are not harmoniously connected, and the video playback is not smooth.

Optionally, in this embodiment of the present invention, the following steps may also be performed before the foregoing sub-step (2a):

Sub-step (2c): determine the third ratio between the fourth number of the second target pictures included in the video clip and the total number of pictures in the video clip; the second target picture is the target information indicating that the picture parameters are different. A target image that meets the preset requirements.

In this step, first, according to the target information of each target picture in the video clip, a target picture whose picture parameters do not meet the preset requirements can be determined, and then a second target picture is obtained. For the specific determination method, reference may be made to the descriptions in the foregoing related steps, which will not be repeated here. Then, the number of the determined second target pictures may be counted to obtain a fourth number. Next, a ratio of the fourth number to the total number of pictures in the video clip may be calculated to obtain a third ratio.

Sub-step (2d): determine a third preset ratio threshold and a third preset number threshold.

Sub-step (2e): if the third ratio is greater than the third preset ratio threshold, and/or the fourth number is greater than the third preset number threshold, discard the video clip.

In this embodiment of the present invention, the third preset ratio threshold and the third preset quantity threshold may be set according to actual requirements. If the third ratio is greater than the third preset ratio threshold, it may be considered that the low-quality pictures in the video clip occupy the proportion of If the fourth number is greater than the third preset number threshold, it can be considered that the number of low-quality pictures in the video clip is large, and it can be determined that the overall quality of the video clip is poor. Therefore, the video clip can be directly discarded. video clips.

At the same time, the overall quality of the video clip is measured in terms of the proportion of low-quality second target pictures and the specific quantity, which can avoid the existence of multiple second targets in the video clip due to the large number of pictures in the video clip. The situation where the picture is not recognized, and avoid the situation that the number of second target pictures is subjectively small, but the second target picture occupies a large proportion in the video clip but is not recognized, so as to ensure that the video with poor quality is determined. Fragment accuracy.

It should be noted that, in this embodiment of the present invention, the video clips whose third ratio is greater than the third preset ratio threshold, and/or the fourth number is greater than the third preset number threshold may also be displayed to the user, and the user receives the video clips for these video clips. select operation, and then delete the video clip selected by the selection operation to ensure the flexibility of user operation.

In the embodiment of the present invention, by determining the proportion or specific number of second target pictures with poor quality appearing in each video segment, the proportion or specific number of the second target pictures is relatively high, that is, the When the overall quality of the video clip is poor, the video clip is directly discarded, thereby reducing the computation amount of subsequent related operations to a certain extent and saving processing resources. It should be noted that, the above-mentioned sub-steps may be executed before sub-step (2a), or may be executed after sub-step (2a). Optionally, it can be performed before sub-step (2a), so that unnecessary culling operations can be avoided, and processing resources can be saved to a greater extent.

Optionally, in this embodiment of the present invention, the above-mentioned step of determining the third preset ratio threshold and the third preset quantity threshold may include: sub-step (2d1): determining, according to the video clip template corresponding to the video to be processed, the third preset ratio threshold and the third preset quantity threshold corresponding to the video clip templates; wherein, the third preset ratio thresholds corresponding to different video clip templates are different, and different video clip templates correspond to The third preset number threshold is different, and the third preset ratio threshold and the third preset number threshold are related to the video content type corresponding to the video clip template.

The video processing method in the embodiment of the present invention can be applied to an automatic video editing scene. Generally, in order to improve the effect of video editing, it is often necessary to remove waste films (pictures with lower quality) in the video. Therefore, based on the video processing method in the embodiment of the present invention, waste pieces in the video clip can be eliminated, so as to improve the efficiency of the video clip and the subsequent viewing experience. Meanwhile, compared with the manual detection and rejection method, the automatic rejection method provided by the embodiments of the present invention can be performed online in real time, thereby reducing the time-consuming of rejection to a certain extent and ensuring the timeliness of processing. By way of example, taking the target parameters including the degree of blur, the degree of shaking, and the degree of exposure as an example, FIG. 2 is a schematic diagram of an editing process provided by an embodiment of the present invention. As shown in FIG. 2 , a user may first select a material (video to be processed), and then perform frame extraction according to a preset frame rate to obtain a picture sequence composed of target pictures. Then, through the CNN network, the fuzzy label information, jitter label information and exposure label information of the target image are determined. Select the clip template, select the clip suitable for clipping. The clips suitable for editing may be the remaining video clips after being discarded through the above sub-step (2e).

Further, the video clip template corresponding to the video to be processed may be selected from optional clip templates. The optional editing template may be preset in the electronic device and may be used for video editing. For example, the optional editing template may include a sports template, a gourmet template, a dynamic template, and the like. When selecting, it may be selected according to the user's preference, for example, the editing template commonly used by the user may be determined according to the user's historical usage record. Then, the commonly used editing template is determined as the video editing template corresponding to the video to be processed. Alternatively, the optional editing template may be directly displayed to the user, and the optional editing template selected by the user is determined as the video editing template corresponding to the video to be processed. Further, the third preset ratio threshold and the third preset quantity threshold corresponding to the video clip template corresponding to the video to be processed may be searched in the preset correspondence between the video clip template and the threshold.

In order to improve the visual effect of the video, when editing the video, the video is often edited according to the video editing template. Among them, corresponding to different video content types, the editing methods of the video editing templates are often different. For example, the corresponding video content type is dynamic content, such as a video editing template that highlights the motion process of the moving subject in the video, such as motion template and dynamic template. When using these templates for editing, the focus is often on the coherence of the video screen. Since the content in such video images is often in motion, it is difficult for users to perceive low-quality images. For static content corresponding to the type of video content, for example, a video editing template that highlights the appearance of a static subject in a video, such as a gourmet template, when using these templates for editing, the focus is often on the appearance details of the subject in the video screen. In this case The motion range of the content in the video picture is often small, so it is easier for the user to perceive that there are more low-quality pictures. Correspondingly, in the embodiment of the present invention, different thresholds may be set for different video editing templates according to the picture characteristics of each video editing template. For example, in the case where the video content type corresponding to the video clip template is dynamic content, a higher third preset ratio threshold and a third preset number threshold are set, and in the case where the video content type corresponding to the video clip template is dynamic content, A lower third preset ratio threshold and third preset quantity threshold are set.

In the embodiment of the present invention, different third preset ratio thresholds and third preset quantity thresholds are correspondingly set for different video editing templates according to the video content types corresponding to the video editing templates, and when editing, according to the currently used Video clip template, select the corresponding threshold to filter video clips, and then to a certain extent, the screening operation can be more adapted to the current clipping needs, thereby improving the effect of clip screening.

FIG. 3 is a block diagram of a video processing apparatus provided by an embodiment of the present invention. The apparatus may include: a memory 301 and a processor 302 . The memory 301 is used to store program codes. The processor 302 calls the program code, and when the program code is executed, is configured to perform the following operations: for the target picture in the video to be processed, determine the target information corresponding to the picture parameter of the target picture; The target information is used to represent whether the picture parameters of the target picture meet the preset requirements; according to the target information of the target picture contained in the video to be processed, the video to be processed is screened; wherein, the screening process It includes deleting a target picture that does not meet the preset requirements or one of the video clips to which the target picture belongs; and generating a target video according to the video to be processed after screening. Specifically, for the specific implementation process of each operation performed by the processor 302, reference may be made to the relevant descriptions in the foregoing method embodiments, which will not be repeated here.

Optionally, the picture parameters include one or more of a blur degree, a shake degree, an exposure degree, and a color change degree of the picture. Optionally, the processor 302 is further configured to: segment the video to be processed to obtain multiple video segments; for each of the video segments, according to the target information of the target picture included in the video segment, Screening processing is performed on the video clips. Optionally, the processor 302 is further configured to: for the target picture in the video to be processed, before determining the target information corresponding to the picture parameter of the target picture, determine the target picture according to the pictures included in the video to be processed. The target picture is extracted; the target picture is extracted from the video to be processed, and timestamp information is added to each of the target pictures; the timestamp information is used to characterize the order of the target picture in the target picture; For the target picture in the video to be processed, after determining the target information corresponding to the picture parameter of the target picture, the target picture is combined into the video to be processed according to the timestamp information of the target picture. Optionally, the processor 302 is further configured to: in the case that the total number of pictures included in the to-be-processed video is greater than a first preset number threshold, extract data from the to-be-processed video according to a preset frame rate. Select m frames of pictures as the target pictures; the m is not greater than the first preset number threshold; when the total number is not greater than the first preset number threshold, the video to be processed is All pictures contained in , as the target picture. Optionally, the processor 302 is further configured to: determine the first ratio of the first number of pixels with a fuzzy confidence level greater than a preset confidence threshold in the target picture to the total number of pixels in the target picture; According to the magnitude relationship between the first ratio and the first preset ratio threshold, the fuzzy label information is determined; the fuzzy label information is used to indicate whether the fuzzy degree of the target picture is normal; and/or, the target image is The picture is used as the input of the first preset classification model, and according to the output category of the first preset classification model, the shaking label information is determined; the first preset classification model is used for classifying pictures according to whether the shaking degree is normal; and/ Or, taking the target picture as the input of the second preset classification model, and determining the exposure label information according to the output category of the second preset classification model; the second preset classification model is used to determine whether the exposure level is normal or not Perform picture classification; and/or, determine the second ratio of the second number of pixels whose color values exceed the preset color value range in the target picture to the total number of pixels; according to the second ratio and the second predetermined The magnitude relationship between the ratio thresholds is set to determine the color difference label information; the color difference label information is used to represent whether the color change degree of the target picture is normal.

Optionally, the processor 302 is further configured to: determine a third number of first target pictures in the video clip; the first target picture is target information indicating that the picture parameters do not meet the preset requirements and are continuous The target pictures that appear; if the third number is greater than the second preset number threshold, remove n of the first target pictures; the n is less than the third number. Optionally, the processor 302 is further configured to: add a picture transition effect to the remaining first target picture; wherein, the picture transition effect includes right-out, left-in, spin-out, spin-in, and fade-in and fade-out one or more of. Optionally, the processor 302 is further configured to: determine a third ratio between the fourth number of second target pictures included in the video clip and the total number of pictures in the video clip; the second target The picture is a target picture whose target information indicates that the picture parameters do not meet the preset requirements; determine a third preset ratio threshold and a third preset number threshold; if the third ratio is greater than the third preset ratio threshold, and /or, if the fourth quantity is greater than the third preset quantity threshold, the video clip is discarded.

Optionally, the processor 302 is further configured to: determine, according to the video clip template corresponding to the video to be processed, the third preset ratio threshold and the third preset number corresponding to the video clip template threshold; wherein the third preset ratio threshold corresponding to different video clip templates is different, the third preset number threshold corresponding to different video clip templates is different, the third preset ratio threshold and the third preset ratio threshold The three preset quantity thresholds are related to the video content type corresponding to the video clip template. Optionally, the processor 302 is further configured to: determine a third target picture whose all picture parameters do not meet the preset requirements; delete the third target picture; wherein the preset requirements include the One or more of normal blur, normal jitter, normal exposure, and normal color variation. Optionally, the processor 302 is further configured to: combine the screened and processed video clips to obtain the target video. Optionally, the processor 302 is further configured to: for the target picture in the video to be processed, before determining the target information corresponding to the picture parameter of the target picture, receive the user's selection operation on the optional video; The optional video selected by the selection operation is determined to be the video to be processed; the optional video is the video stored in the electronic device; after the target video is generated according to the screened video to be processed, the target video is displayed to the user. video.

To sum up, the video processing apparatus provided by the embodiment of the present invention determines, for the target picture in the video to be processed, target information corresponding to the picture parameters of the target picture, wherein the target information is used to indicate whether the picture parameters of the target picture meet the predetermined requirements. Setting requirements, according to the target information of the target picture contained in the video to be processed, the video to be processed is screened, wherein the screening process includes deleting the target picture that does not meet the preset requirements or one of the video clips to which the target picture belongs, and finally, Generate a target video according to the to-be-processed video after screening. In this way, by automatically screening the pictures in the video according to the picture parameters of the pictures during the video processing, the screening cost can be reduced to a certain extent and the screening efficiency can be improved. Further, an embodiment of the present invention further provides a movable device, where the movable device includes a video capture device, and the movable device is configured to capture a video to be processed through the video capture device. According to the above video processing The method processes the video to be processed. Optionally, the movable device is a drone and/or an unmanned vehicle. Further, an embodiment of the present invention also provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, each step in the above video processing method is implemented, and can To achieve the same technical effect, in order to avoid repetition, details are not repeated here. The device embodiments described above are only illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed over multiple network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort. Various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor may be used in practice to implement some or all of the functions of some or all of the components in the computing processing device according to the embodiments of the present invention. The present invention can also be implemented as apparatus or apparatus programs (eg, computer programs and computer program products) for performing part or all of the methods described herein. Such a program implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such signals may be downloaded from Internet sites, or provided on carrier signals, or in any other form.

For example, FIG. 4 is a block diagram of a computing processing device provided by an embodiment of the present invention. As shown in FIG. 4 , FIG. 4 shows a computing processing device that can implement the method according to the present invention. The computing processing device traditionally includes a processor 710 and a computer program product or computer readable medium in the form of a memory 720 . The memory 720 may be electronic memory such as flash memory, EEPROM (electrically erasable programmable read only memory), EPROM, hard disk, or ROM. The memory 720 has storage space 730 for program code for performing any of the method steps in the above-described methods. For example, the storage space 730 for program codes may include various program codes for implementing various steps in the above methods, respectively. These program codes can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards or floppy disks. Such computer program products are typically portable or fixed storage units as described with reference to FIG. 5 . The storage unit may have storage segments, storage spaces, etc. arranged similarly to the memory 720 in the computing processing device of FIG. 4 . The program code may, for example, be compressed in a suitable form. Typically, the storage unit includes computer readable code, ie code readable by a processor such as 710 for example, which when executed by a computing processing device, causes the computing processing device to perform each of the methods described above. step. The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments may be referred to each other. Reference herein to "one embodiment," "an embodiment," or "one or more embodiments" means that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment of the present invention. Also, please note that instances of the phrase "in one embodiment" herein are not necessarily all referring to the same embodiment. In the description provided herein, numerous specific details are set forth. It will be understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. do not denote any order. These words can be interpreted as names.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that it can still be The technical solutions described in the foregoing embodiments are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

A video processing method, characterized in that the method comprises:

For the target picture in the video to be processed, determine the target information corresponding to the picture parameter of the target picture; the target information is used to represent whether the picture parameter of the target picture meets the preset requirements;

Screening the video to be processed according to the target information of the target picture contained in the video to be processed; wherein, the screening process includes deleting target pictures that do not meet the preset requirements or the target pictures to which the target pictures belong. one of the video clips;

Generate a target video according to the to-be-processed video after screening.
The method according to claim 1, wherein the picture parameters include one or more of the degree of blurring, the degree of shaking, the degree of exposure, and the degree of color change of the picture.
The method according to claim 1 or 2, characterized in that, performing screening processing on the to-be-processed video according to target information of a target picture included in the to-be-processed video comprises:

segmenting the to-be-processed video to obtain a plurality of video clips;

For each of the video segments, the video segment is screened according to the target information of the target picture contained in the video segment.
The method according to any one of claims 1 to 3, wherein, for the target picture in the video to be processed, before determining the target information corresponding to the picture parameter of the target picture, the method further comprises: according to the The picture contained in the video to be processed is determined, and the target picture is determined; the target picture is extracted from the video to be processed, and timestamp information is added to each of the target pictures; the timestamp information is used to represent the the order of the target picture in the target picture;

For the target picture in the video to be processed, after determining the target information corresponding to the picture parameter of the target picture, the method further includes: combining the target picture into the target picture according to the timestamp information of the target picture. Video to be processed.
The method according to claim 4, wherein the determining the target picture according to the picture included in the to-be-processed video comprises:

When the total number of pictures included in the video to be processed is greater than the first preset number threshold, select m frames of pictures from the video to be processed according to a preset frame rate as the target picture; the m not greater than the first preset number threshold;

Under the condition that the total number is not greater than the first preset number threshold, all pictures included in the video to be processed are used as the target picture.
The method according to claim 1 or 2, wherein, for the target picture in the video to be processed, determining the target information corresponding to the picture parameter of the target picture, comprising:

Determine the first ratio of the first number of pixels whose fuzzy confidence is greater than a preset reliability threshold in the target picture to the total number of pixels in the target picture; according to the ratio between the first ratio and the first preset ratio threshold The size relationship between the two is to determine the fuzzy label information; the fuzzy label information is used to represent whether the fuzzy degree of the target picture is normal;

And/or, taking the target picture as the input of the first preset classification model, and determining the jitter label information according to the output category of the first preset classification model; the first preset classification model is used to determine the jitter label information according to the degree of jitter Whether the image classification is performed normally;

And/or, the target picture is used as the input of the second preset classification model, and the exposure label information is determined according to the output category of the second preset classification model; the second preset classification model is used for according to the exposure degree. Whether image classification is performed normally;

And/or, determining the second ratio between the second number of pixels whose color value exceeds the preset color value range in the target picture and the total number of pixels; according to the ratio between the second ratio and the second preset ratio threshold; The size relationship between the two is determined, and the color difference label information is determined; the color difference label information is used to represent whether the color change degree of the target picture is normal.
The method according to claim 3, wherein the filtering of the video clip according to the target information of the target picture included in the video clip comprises:

determining the third quantity of the first target picture in the video clip; the first target picture is the target picture whose target information indicates that the picture parameter does not meet the preset requirement and appears continuously;

If the third number is greater than the second preset number threshold, remove n of the first target pictures; the n is less than the third number.
The method according to claim 7, wherein after removing the n first target pictures, the method further comprises:

adding a picture transition effect to the remaining first target pictures;

Wherein, the picture transition effects include one or more of right-out, left-in, spin-out and spin-in, and fade-in and fade-out.
The method according to claim 7, wherein the method further comprises:

Determine the third ratio of the fourth number of second target pictures included in the video clip to the total number of pictures in the video clip; the second target picture is target information indicating that the picture parameters do not meet the preset requirements the target image;

determining a third preset ratio threshold and a third preset number threshold;

If the third ratio is greater than the third preset ratio threshold, and/or the fourth number is greater than the third preset number threshold, discarding the video segment.
The method according to claim 9, wherein the determining the third preset ratio threshold and the third preset quantity threshold comprises:

According to the video clip template corresponding to the video to be processed, determine the third preset ratio threshold and the third preset number threshold corresponding to the video clip template;

The third preset ratio threshold corresponding to different video clip templates is different, the third preset number threshold corresponding to different video clip templates is different, the third preset ratio threshold and the third preset ratio threshold are different. The number threshold is set to be related to the video content type corresponding to the video clip template.
The method according to claim 4, wherein, before incorporating the target picture into the to-be-processed video according to the timestamp information of the target picture, the method further comprises:

determining a third target picture for which all picture parameters do not meet the preset requirements;

delete the third target picture;

Wherein, the preset requirements include one or more of the normal blur degree, normal jitter degree, normal exposure degree and normal color change degree.
The method according to claim 3, wherein generating the target video according to the to-be-processed video after screening and processing comprises:

The target video is obtained by merging the screened video clips.
The method according to claim 1, wherein, for the target picture in the to-be-processed video, before determining the target information corresponding to the picture parameter of the target picture, the method further comprises: receiving user feedback on the optional video the selection operation; determine the optional video selected by the selection operation as the to-be-processed video; the optional video is the video stored in the electronic device;

After generating the target video according to the screened video to be processed, the method further includes: displaying the target video to the user.
A video processing device, characterized in that the device includes a memory and a processor;

the memory for storing program codes;

The processor calls the program code, and when the program code is executed, is configured to perform the following operations:

For the target picture in the video to be processed, determine the target information corresponding to the picture parameter of the target picture; the target information is used to represent whether the picture parameter of the target picture meets the preset requirements;

Screening the to-be-processed video according to the target information of the target picture contained in the to-be-processed video; wherein the screening process includes deleting the target picture that does not meet the preset requirements or the target picture to which the target picture belongs. one of the video clips;

Generate a target video according to the to-be-processed video after screening.
A movable device, characterized in that the movable device includes a video capture device, and the movable device is configured to capture a video to be processed through the video capture device, according to any one of claims 1 to 13 The video processing method processes the to-be-processed video.
The method according to claim 14, wherein the movable device is an unmanned aerial vehicle and/or an unmanned vehicle.
A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the following operations are implemented:

For the target picture in the video to be processed, determine the target information corresponding to the picture parameter of the target picture; the target information is used to represent whether the picture parameter of the target picture meets the preset requirements;

Screening the to-be-processed video according to the target information of the target picture contained in the to-be-processed video; wherein the screening process includes deleting the target picture that does not meet the preset requirements or the target picture to which the target picture belongs. one of the video clips;

Generate a target video according to the to-be-processed video after screening.