Television video search method and device, television and storage medium

Abstract

The present invention discloses a method and device for searching television videos, a television, and a storage medium, relating to the field of smart television technology. The method comprises: obtaining an input image; performing image recognition on the input image to obtain a target object; and searching pre-stored information for a video clip corresponding to the target object based on the current channel of the television to obtain a video search result. The present invention achieves the purpose of directly using images without the need for text input, thus resolving the problem of low intelligence in video search on existing smart televisions.

Description

Technical Field

[0001] The present invention relates to the technical field of smart televisions, and in particular to a television video search method and device, a television, and a storage medium. Background Art

[0002] With the continuous development of smart home appliances, more and more Internet TVs and smart TVs have emerged. Smart TVs are TV products that meet the diverse and personalized needs of users. Nowadays, the processors of smart TVs are becoming more and more powerful, the memory space is becoming more and more sufficient, and the functions are also becoming more complex.

[0003] Currently, the most user-friendly feature for watching TV channels on smart TVs is limited to limited-time playback. Installed video apps rely solely on the app's own functionality and cannot be expanded by the TV itself. When users need to find a video, they can only enter text in the video app to search for it. Therefore, existing smart TVs lack personalized video search capabilities and lack a high level of intelligence. Summary of the Invention

[0004] The main purpose of the present invention is to provide a television video search method and device, a television and a storage medium, aiming to solve the technical problem that existing smart televisions have a low degree of intelligence in video search.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a method for searching for television videos, the method comprising:

[0007] Get the input image;

[0008] Performing image recognition on the input image to obtain a target object;

[0009] According to the current channel of the television, the video clip corresponding to the target object is searched in the pre-stored information to obtain a video search result.

[0010] Optionally, in the above-mentioned television video search method, the step of searching for a video clip corresponding to the target object in pre-stored information according to the current channel of the television to obtain a video search result specifically includes:

[0011] When it is detected that the current channel of the television is a television channel, obtaining a first video clip corresponding to the target object according to a pre-stored parameter list; and / or

[0012] When it is detected that the current channel of the television is a video application channel, obtaining a second video clip corresponding to the target object according to video application storage information; and / or

[0013] When it is detected that the current channel of the television is a video application channel, a keyword search is performed in the video application to obtain a third video clip corresponding to the target object, wherein the keyword is obtained based on the target object;

[0014] The first video segment, the second video segment and / or the third video segment are aggregated to obtain a video search result.

[0015] Optionally, in the above-mentioned television video search method, when it is detected that the current channel of the television is a television channel, the step of obtaining the first video clip corresponding to the target object according to a pre-stored parameter list specifically includes:

[0016] When it is detected that the current channel of the television is a television channel channel, obtaining the television channel containing the target object according to a pre-stored television channel list;

[0017] A keyword search is performed according to the parameter list of the television channel to obtain a corresponding first video clip.

[0018] Optionally, in the above-mentioned television video search method, before the step of obtaining the television channel containing the target object according to a pre-stored television channel list when the current channel of the television is detected to be a television channel channel, the method further includes:

[0019] According to the parameter list of each television channel, a summary is performed to obtain a television channel summary list, wherein the parameter list includes at least one of a program name list, a program content list and a channel label list.

[0020] Optionally, in the above-mentioned television video search method, before the step of aggregating the parameter list of each television channel to obtain a total television channel list, the method further includes:

[0021] When it is detected that the current channel of the television is a television channel and a television program is being played, the television picture of the current television channel is captured according to a preset time interval to obtain a screenshot image;

[0022] Performing image recognition on the screenshot image to obtain all objects in the screenshot image;

[0023] Counting all objects in all the screenshots within a preset time period to obtain label information of the video clips corresponding to the preset time period;

[0024] A channel label list of the current television channel is obtained according to the video clip and the label information.

[0025] Optionally, in the above-mentioned television video search method, after the step of counting all objects in all screenshot images within a preset time period, the method further includes:

[0026] When an operation instruction requesting to edit a tag is obtained from the user, the obtained statistical results of all objects are displayed on the display interface of the television, wherein the statistical results include the frequency of occurrence of the objects;

[0027] According to the user's operation instruction of adding or deleting an object, the object is added or deleted in the statistical result accordingly to obtain the edited statistical result;

[0028] According to the preset filtering conditions, the objects in the edited statistical results are filtered to obtain the label information of the video clips corresponding to the preset time period.

[0029] Optionally, in the above-mentioned television video search method, the object includes at least one of an object, a person, and a scene;

[0030] The step of performing image recognition on the input image to obtain the target object specifically includes:

[0031] Performing image recognition on the input image using image recognition technology to obtain all objects in the input image and displaying them on the display interface of the television;

[0032] When an operation instruction for selecting an object is obtained from the user, the object selected by the user is used as the target object.

[0033] In a second aspect, the present invention provides a television video search device, the device comprising:

[0034] An image acquisition module, used to acquire an input image;

[0035] An image recognition module, configured to perform image recognition on the input image to obtain a target object;

[0036] The video search module is used to search for the video clip corresponding to the target object in the pre-stored information according to the current channel of the television, and obtain the video search result.

[0037] In a third aspect, the present invention provides a television, comprising a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the television video search method as described above is implemented.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program can be executed by one or more processors to implement the television video search method as described above.

[0039] The above one or more technical solutions provided by the present invention may have the following advantages or at least achieve the following technical effects:

[0040] The present invention proposes a television video search method and device, a television, and a storage medium. The method obtains an input image, performs image recognition on the input image, obtains a target object, and then searches for a video clip corresponding to the target object in pre-stored information for the current channel of the television to obtain a video search result. This method achieves the purpose of directly using images to search for videos without entering text, and video searches can be performed for different channels of the television, which not only expands the search range but also improves the intelligence level of the television. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these provided drawings without paying any creative work.

[0042] Figure 1 1 is a flow chart of a first embodiment of a method for searching for television videos according to the present invention;

[0043] Figure 2 A schematic diagram of the hardware structure of a television according to the present invention;

[0044] Figure 3 Schematic diagram of the flow of a second embodiment of the television video search method of the present invention;

[0045] Figure 4 FIG. 1 is a schematic diagram of functional modules of a first embodiment of a television video search device according to the present invention.

[0046] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0047] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0048] It should be noted that in the present invention, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "includes..." does not exclude the presence of other identical elements in the process, method, article or system comprising the element. In addition, the meaning of "and / or" appearing throughout the text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme in which both A and B are satisfied.

[0049] In the present invention, if there are descriptions involving "first", "second", etc., the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of such features. In addition, in the present invention, suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present invention and have no specific meaning in themselves. Therefore, "module", "component" or "unit" can be used interchangeably. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances. In addition, the technical solutions of each embodiment can be combined with each other, but it must be based on the ability of ordinary technicians in this field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0050] An analysis of existing technologies reveals that there are two main ways for users to watch TV programs: one is to jump to TV channels one by one in sequence, and the other is to select the corresponding program video in the video application channel. Some smart TVs have also added more user-friendly functions based on this. There are mainly two solutions:

[0051] 1. When watching a TV channel, you can replay the current channel for a limited time. For example, you can replay a program that was broadcast on the current channel an hour ago. When you need to watch a program, you need to know which TV channel it is on, jump to that channel, select Replay, and adjust the progress bar accordingly. Not only is the replay time limited, but the operation is also relatively complicated and has significant limitations.

[0052] 2. When watching a video application channel, when the user needs to find a video, they enter text in the video application to search for the video. Since the TV uses a remote control for human-computer interaction, when entering text, they can only enter it through the soft keyboard, which is very inconvenient.

[0053] Therefore, existing smart TVs cannot provide users with a more personalized video search function, and have a low level of intelligence in video search.

[0054] In view of the technical problem that the smart TV in the prior art has a low degree of intelligence, the present invention provides a method for searching TV videos. The overall idea is as follows:

[0055] Acquire an input image; perform image recognition on the input image to obtain a target object; and search for a video clip corresponding to the target object in pre-stored information according to the current channel of the television to obtain a video search result.

[0056] Through the above technical solution, an input image is obtained, image recognition is performed on the input image, and after the target object is obtained, the video clip corresponding to the target object is searched in the pre-stored information for the current channel of the TV to obtain a video search result; the purpose of directly using images to search for videos without entering text is achieved, and video searches can be performed for different channels of the TV, which not only expands the search range but also improves the intelligence level of the TV.

[0057] Example 1

[0058] Reference Figure 1 The flowchart of FIG. 1 is a flowchart of a first embodiment of a method for searching for television videos according to the present invention, wherein the method is applied to a television. The television is a smart television or a set-top box capable of achieving network connection.

[0059] like Figure 2 1 is a schematic diagram of the hardware structure of a television, which may include: a processor 1001 , such as a CPU (Central Processing Unit), a communication bus 1002 , a user interface 1003 , a network interface 1004 , and a memory 1005 .

[0060] Those skilled in the art will understand that Figure 2 The hardware structure shown in the figure does not constitute a limitation on the television of the present invention, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0061] Specifically, the communication bus 1002 is used to implement connection and communication between these components;

[0062] The user interface 1003 is used to connect to the client and perform data communication with the client. The user interface 1003 may include an output unit, such as a display screen, and an input unit, such as a keyboard and a camera. Optionally, the user interface 1003 may also include other input / output interfaces, such as a standard wired interface and a wireless interface, such as a wirelessly connected TV remote control.

[0063] The network interface 1004 is used to connect to the backend server and perform data communication with the backend server. The network interface 1004 may include an input / output interface, such as a standard wired interface or a wireless interface, such as a Wi-Fi interface.

[0064] The memory 1005 is used to store various types of data. These data may include, for example, instructions for any application or method in the television, as well as application-related data. The memory 1005 may be a high-speed RAM memory or a stable memory such as a disk memory. Optionally, the memory 1005 may also be a storage device independent of the processor 1001.

[0065] For details, please refer to Figure 2 , the memory 1005 may include an operating system, a network communication module, a user interface module and a computer program, wherein the network communication module is mainly used to connect to the server and perform data communication with the server;

[0066] The processor 1001 is configured to call the computer program stored in the memory 1005 and perform the following operations:

[0067] Get the input image;

[0068] Performing image recognition on the input image to obtain a target object;

[0069] According to the current channel of the television, the video clip corresponding to the target object is searched in the pre-stored information to obtain a video search result.

[0070] Based on the above TV, the following Figure 1 The flowchart shown in FIG. 1 describes in detail the TV video search method of this embodiment. The method may include the following steps:

[0071] Step S10: Obtain an input image.

[0072] Specifically, the input image is an image related to the object the user wants to search for, such as an image of an object, person, or scene. The input image can be obtained by taking a photo with the TV's built-in camera, or by transmitting the image to the TV via a wired or wireless connection. The TV then stores the image locally and retrieves the input image from the local storage space.

[0073] Step S20: performing image recognition on the input image to obtain a target object.

[0074] Specifically, by using image recognition technology to perform object recognition on the input image and obtain all objects in the image, the television can automatically determine the target object, or the user can select one of all objects as the target object.

[0075] Step S30: searching for a video clip corresponding to the target object in pre-stored information according to the current channel of the television to obtain a video search result.

[0076] Specifically, the TV channel includes TV channel channels, such as CCTV1, CCTV5, CCTV10 and other TV channels, and also includes video application channels, such as applications provided specifically for TV by various video operators. The pre-stored information may include a parameter list of the TV channel and video application storage information. The video application storage information may include local storage content of the video application, content provided by the video application server, and so on. The search method includes using the parameter list of the TV channel to search for videos containing the target object in the locally stored TV channel program videos, or automatically searching for videos containing the target object in the locally stored video application cache videos, or searching a large number of videos stored on the server side of the video application through the network to find videos containing the target object. You can select a corresponding search method for the current channel, or you can select multiple feasible search methods for the current channel to obtain matching video search results, and present the results to the user in the form of a list or chart.

[0077] The television video search method provided in this embodiment obtains an input image, performs image recognition on the input image, obtains a target object, and then searches for a video clip corresponding to the target object in pre-stored information for the current channel of the television to obtain a video search result. This method achieves the purpose of performing video search directly using images without the need for text input. Moreover, video search can be performed for different channels of the television, which not only expands the search range but also improves the intelligence level of the television.

[0078] Example 2

[0079] Based on the same invention concept, Figure 3 , a second embodiment of the television video search method of the present invention is proposed, and the television video search method is applied to television.

[0080] The following combination Figure 3 The flowchart shown in FIG. 1 describes in detail the TV video search method of this embodiment. The method may include the following steps:

[0081] Step S100: Acquire an input image.

[0082] Specifically, the input image can be obtained by taking a photo with a TV camera, or by retrieving a locally stored photo. This embodiment uses a photo of a basketball star holding a basketball as an example for explanation.

[0083] Step S200: performing image recognition on the input image to obtain a target object.

[0084] Specifically, the object includes at least one of an object, a person, and a scene; and step S200 may include:

[0085] Step S210: performing image recognition on the input image using image recognition technology to obtain all objects in the input image and display them on the display interface of the television;

[0086] Specifically, image recognition technology refers to the technology of object recognition in images to identify targets and objects of various different patterns. At present, image recognition technology has gradually matured and is not specifically limited here. It is based on the ability to recognize all objects contained in the input image. Among them, the object category can be objects, such as flowers, sports equipment, animals, etc., or people, such as some famous people, or even scenes, such as landscapes, sports fields, etc. These object categories can be recognized at the same time, or only one category can be recognized. It can be displayed on the TV display interface, either in the form of image frame selection or in the form of a list.

[0087] In this embodiment, the input image is recognized, and the objects obtained include a basketball and a basketball star. The image can then be displayed on a display interface, with the basketball and the face of the basketball star marked with two frames for user selection.

[0088] Step S220: When an operation instruction for selecting an object is obtained, the object selected by the user is used as a target object.

[0089] Specifically, for all objects of the input image displayed on the display interface, the user controls the display interface through the remote control, selects one of the objects, generates a corresponding operation instruction, and the TV responds to the instruction with the object selected by the user as the target object.

[0090] In this embodiment, with respect to the display interface that marks the basketball and the face of the basketball star, the frame that marks the face of the basketball star is selected, and the basketball star is used as the target object.

[0091] Step S300: searching for a video clip corresponding to the target object in pre-stored information according to the current channel of the television to obtain a video search result.

[0092] Specifically, step S300 may include:

[0093] Step S310: when it is detected that the current channel of the television is a television channel, obtaining a first video clip corresponding to the target object according to a pre-stored parameter list; and / or

[0094] Step S320: when it is detected that the current channel of the television is a video application channel, obtaining a second video clip corresponding to the target object according to the video application storage information; and / or

[0095] Step S330: When it is detected that the current channel of the television is a video application channel, a keyword search is performed in the video application to obtain a third video clip corresponding to the target object, wherein the keyword is obtained based on the target object.

[0096] Specifically, you can choose any one of the three search methods mentioned above, that is, when in the TV channel channel, you can only obtain the first video clip, and when in the video application channel, you can only obtain the second video clip or only obtain the third video clip; you can also search in a mixed manner, for example, you can first obtain the first video clip in the TV channel channel, and then obtain the second video clip and the third video clip in the video application channel.

[0097] Specifically, step S310 may include:

[0098] Step S311: when it is detected that the current channel of the television is a television channel channel, obtaining the television channel containing the target object according to a pre-stored television channel list;

[0099] Specifically, when the current TV channel is identified as a TV channel channel, a search is performed in the pre-stored TV channel master table to obtain all TV channels containing the target object. The list can be displayed to the user, and the user can select a TV channel to jump to. Alternatively, the user can automatically jump to the TV channel containing the target object, for example, directly jump to the TV channel closest to the current TV channel. The automatic jump rules can be set according to actual conditions. The TV channel master table includes the content of each TV channel in each time period. The time period can be a fixed interval set directly, such as one hour as a time period, or it can be an interval based on a program, for example, one program as a time period, ignoring advertisements inserted between programs. The content can include program names, program content introductions, custom channel labels, etc.

[0100] In this embodiment, assume that the current TV channel is a TV channel, and specifically CCTV4 is being watched. This TV channel is ranked fourth in the TV channel master list. When the user determines that the target object is a basketball star, the TV channels that contain this basketball star are determined to be CCTV1, CCTV5, and CCTV Sports according to the TV channel master list. This result is then displayed to the user. The user can manually select CCTV Sports to directly jump from the current CCTV4 to CCTV Sports, or the user can automatically jump to the TV channel closest to CCTV4. Assuming that CCTV1 is ranked first, CCTV5 is ranked fifth, and CCTV Sports is ranked fiftieth in the TV channel master list, in this embodiment, the user can jump directly to CCTV5.

[0101] Step S312: performing a keyword search according to the parameter list of the television channel to obtain a corresponding first video clip.

[0102] Specifically, the parameter list includes at least one of a program name list, a program content list, and a channel label list. Specifically, it includes the program name, program content, or channel label for each time period of the television channel. The search for the first matching video clip can be performed based on the program name list only, the channel label list only, or the program name list first, then the program content list, and finally the channel label. The specific selection can be made based on actual circumstances.

[0103] In this embodiment, the search is performed in the parameter list of CCTV5, for example, in its program name list and channel label list. When the target object is a basketball star obtained in step S200, the name of the basketball star is obtained as a keyword, for example, basketball star A. Assume that the program name list of the TV channel is as shown in Table 1:

[0104] Table 1

[0105] Video clips Program Name 2021.01.01 18:00-19:30 The legendary story of basketball star A …… …… 2021.01.02 15:00-16:30 First basketball game 2021.01.02 16:40-18:10 Basketball Game 2 …… …… 2021.01.07 12:00-13:00 Basketball News

[0106] According to Table 1, the first video segment can be obtained as the video segment corresponding to the time period of 2021.01.01 18:00-19:30.

[0107] But in fact, basketball star A is also included in the first basketball game and basketball news. Therefore, the program content list of the TV channel can also be used for searching. Assume that the program name list of the TV channel is as shown in Table 2:

[0108] Table 2

[0109]

[0110] According to Table 2, the first video segments can be obtained as video segments corresponding to the two time periods of 2021.01.01 18:00-19:30 and 2021.01.02 15:00-16:30.

[0111] But in fact, basketball news also has basketball star A. Therefore, the channel label list of the TV channel can also be used for search. Assume that the channel label list of the TV channel is as shown in Table 3:

[0112] Table 3

[0113] Video clips Channel Tags 2021.01.01 18:00-19:30 Basketball star A, basketball, football, host... …… …… 2021.01.02 15:00-16:30 Basketball, Basketball Star A, Basketball Star B, ... 2021.01.02 16:40-18:10 Basketball, basketball star C, basketball star D, ... …… …… 2021.01.07 12:00-13:00 Trophy, basketball star A, basketball, basketball star B...

[0114] According to Table 3, we can obtain the first video clips corresponding to the three time periods of 2021.01.01 18:00-19:30, 2021.01.02 15:00-16:30, and 2021.01.07 12:00-13:00. These video clips all contain basketball star A, which is more accurate than searching by program name alone. In actual situations, a parameter list that includes program name, program content, and channel label can also be included as a search pool for a more comprehensive search to prevent the situation where the channel label does not contain the target object, but the program name does.

[0115] More specifically, in the above parameter list, there is no need to create a list for each time period of each channel. It is possible to create and update the corresponding parameter list only for the TV channels watched by the user, or even create and update the corresponding parameter list only for the time period watched by the user, to prevent the video clips corresponding to the time period that the user is not interested in from occupying too many resources.

[0116] Specifically, in step S320, when it is detected that the current channel of the television is a video application channel, a second video clip containing the target object can be obtained based on the browsing history and favorites stored by the video application. Specifically, the browsing history can be locally cached video content or video content selected by the user. Obtaining the second video clip by searching only for keywords in the browsing history and favorites stored by the video application can more accurately determine the user's favorite content and correspondingly search for video clips containing the target object, resulting in a more personalized and accurate search.

[0117] Specifically, in step S330, a keyword search is performed in the video application to obtain a third video clip containing the target object, and the search is performed in all video clips of the video application. This search method is more comprehensive than the search method in step S330.

[0118] Step S340: Summarize the first video clip, the second video clip and / or the third video clip to obtain a video search result.

[0119] Specifically, after the corresponding video clips are found in the above steps, these video clips are summarized and displayed to the user in a list form. When displaying, the video clips can also be sorted, such as by relevance or by time, etc., which is not specifically limited.

[0120] In this embodiment, assuming that the search is only performed on the TV channel, the search results include video clips corresponding to the three time periods of 2021.01.01 18:00-19:30, 2021.01.02 15:00-16:30 and 2021.01.07 12:00-13:00. The above video clips can be sorted by relevance and displayed to the user.

[0121] In one embodiment, before step S311, the method further includes:

[0122] Step S301: when it is detected that the current channel of the television is a television channel and a television program is being played, the television picture of the current television channel is captured according to a preset time interval to obtain a screenshot image.

[0123] Specifically, when a TV program is playing normally, a screenshot of the current TV screen can be taken at intervals to obtain a screenshot image. For example, this embodiment uses watching CCTV5 on January 2, 2021 as an example. The preset time interval is set to 5 minutes. Assuming that the normal TV broadcast period is 14:00-19:30, multiple screenshot images will be obtained accordingly.

[0124] Screen capture is performed for the TV channels being played, and no tags are added to the TV channels that have not been watched. This allows users to more quickly find video clips that have been watched and are related to the content they are looking for.

[0125] Step S302: performing image recognition on the screenshot image to obtain all objects in the screenshot image.

[0126] Specifically, image recognition is performed separately for each screenshot to identify all objects in the screenshot. For example, in this embodiment, the screenshot captured at 3:05 PM contains objects such as a basketball, basketball star A, and basketball star C; the screenshot captured at 3:10 PM contains objects such as a basketball, basketball star A, and basketball star D; and the screenshot captured at 3:15 PM contains objects such as a basketball, basketball star A, and basketball star B. The object categories to be identified can be set according to actual circumstances.

[0127] Step S303: performing statistics on all objects in all the screenshot images within a preset time period to obtain label information of the video clip corresponding to the preset time period.

[0128] Specifically, the preset time period can be a time period set with a fixed duration, such as one hour or two hours, with no interval between each time period, or it can be a program time period set with the duration of a program, in which advertising time can be omitted; when statistics are taken, statistics can be taken according to the frequency of occurrence of the identified objects, or statistics can be taken based on other references, such as the number of times an object appears twice in a row, etc.

[0129] After statistics are collected, keywords that meet the conditions are filtered out according to predetermined rules and used as label information for the video clips corresponding to the time period. For example, in this embodiment, object statistics are performed for the program time periods of 13:00-14:50, 15:00-16:30, 16:40-18:10, 18:13-19:00, and 19:00-19:30 respectively. For each program time period, keywords corresponding to objects that meet preset conditions, such as a frequency of more than 10 times, are used as labels for the video clips corresponding to the program time period. Label sorting can also be sorted by frequency. For example, the objects obtained for each program time period in this embodiment are shown in Table 4:

[0130] Table 4

[0131]

[0132] Statistics are performed on all objects in Table 4 above, specifically counting the frequency of each object's appearance. For example, in Table 4 above, basketball appears 4 times in the 15:00-16:30 program period and 6 times on the CCTV5 channel. After the objects are counted in this manner, they are filtered according to the preset rules. For example, in this embodiment, the filtering condition can be to sort the objects with the highest number of appearances from the highest to the lowest, and then use the objects with the highest number of appearances as the final label information, thereby filtering out the label information that meets the conditions. The label information obtained in this embodiment is shown in Table 5:

[0133] Table 5

[0134] Program time Tag information 14:00-14:50 an athlete, volleyball, a volleyball coach 15:00-16:30 Basketball, Basketball Star A, Basketball Star B 16:40-18:10 Basketball, Basketball Star C, Basketball Star D 18:13-19:00 Football, a football player, a referee 19:00-19:30 Host, studio, news desk

[0135] Step S304: Obtain a channel label list of the current TV channel according to the video clip and the label information.

[0136] Specifically, based on the statistically obtained tag information and the corresponding video clips within the preset time period, a channel tag list of the TV channel is obtained. For example, in this embodiment, the channel tag list shown in Table 6 is obtained:

[0137] Table 6

[0138] Video clips Tag information 2021.01.02 14:00-14:50 an athlete, volleyball, a volleyball coach 2021.01.02 15:00-16:30 Basketball, Basketball Star A, Basketball Star B 2021.01.02 16:40-18:10 Basketball, Basketball Star C, Basketball Star D 2021.01.02 18:13-19:00 Football, a football player, a referee 2021.01.02 19:00-19:30 Host, studio, news desk

[0139] According to the above steps, you can obtain the channel label list of CCTV5. For this list, you can continue to add video clips and corresponding label information during the subsequent viewing of the TV channel. Correspondingly, you can also set the upper limit for saving video and label information. For example, you can set it to only retain video clips watched in the last six months.

[0140] Step S305: Summarize the parameter lists of each television channel to obtain a television channel summary list, wherein the parameter lists include at least one of a program name list, a program content list, and a channel label list.

[0141] Specifically, the channel label list is obtained for the TV channels that are being watched. For the TV channels that are not being watched, the program name list and program content list can be established based on the program names and program contents synchronized by the TV station. This example is not given here. Finally, the program names, program names and label information of all TV channels are summarized to obtain a TV channel summary table, such as shown in Table 7:

[0142] Table 7

[0143]

[0144] For each TV channel, the program name and content of the corresponding video clip in the preset time period are obtained. Based on the programs watched, the corresponding tag information is obtained. For programs not watched, their tag information cannot be obtained. After summarizing, a total TV channel table is obtained.

[0145] The above steps S301 to S305 may be independently executed step branches, thereby achieving the purpose of aggregating data on the watched TV channels and obtaining or updating the TV channel list in real time.

[0146] In another embodiment, step S303 may include:

[0147] Step S303.1: When a user operation instruction to edit a tag is obtained, the obtained statistical results of all objects are displayed on the display interface of the television, wherein the statistical results include the frequency of occurrence of the objects;

[0148] Step S303.2: According to the user's request to add or delete an object, the object is added or deleted in the statistical result accordingly to obtain the edited statistical result;

[0149] Step S303.3: Filter the objects in the edited statistical results according to the preset filtering conditions to obtain the label information of the video clips corresponding to the preset time period.

[0150] Specifically, this embodiment can add a user editing function for the scenarios in Table 4 or Table 5. That is, the user operates the remote control to request tag editing on the TV, and the TV display interface shows the object statistics for the current program time period, including the object's frequency of occurrence. The user can manually add or delete objects for all current objects, and then filter out qualified objects based on preset conditions, thereby obtaining tag information for the video clips viewed in each preset time period.

[0151] It should be noted that after the step of counting all objects in all the screenshot images within the preset time period, steps S303.1 to S303.3 can be executed in sequence to achieve the purpose of summarizing data on the watched TV channels and obtaining or updating the TV channel list in real time. Steps S303.1 to S303.3 can also be independently executed step branches to achieve the purpose of manually editing label information.

[0152] The television video search method provided in this embodiment is based on image recognition technology and object recognition technology, and provides a method for adding tags to television channels and programs, as well as a method for performing corresponding video search based on the total list of television channels obtained after adding tags, thereby improving the hit rate of users finding their favorite programs; multiple program search methods increase the comprehensiveness of video search, and video clip searches are performed based on multiple parameter lists, making the search more accurate and enabling more personalized search, thereby improving the personalized function of television.

[0153] Example 3

[0154] Based on the same invention concept, Figure 4 , a first embodiment of a television video search device of the present invention is proposed, which can be a virtual device applied to a television.

[0155] The following combination Figure 4 The functional module diagram shown in FIG. 1 is a detailed description of the television video search device provided in this embodiment. The device may include:

[0156] An image acquisition module, used to acquire an input image;

[0157] An image recognition module, configured to perform image recognition on the input image to obtain a target object;

[0158] The video search module is used to search for the video clip corresponding to the target object in the pre-stored information according to the current channel of the television, and obtain the video search result.

[0159] Furthermore, the video search module may include:

[0160] a first search submodule, configured to obtain a first video clip corresponding to the target object according to a pre-stored parameter list when detecting that the current channel of the television is a television channel; and / or

[0161] A second search submodule is configured to obtain a second video clip corresponding to the target object according to video application storage information when it is detected that the current channel of the television is a video application channel; and / or

[0162] a third search submodule, configured to, when detecting that the current channel of the television is a video application channel, perform a keyword search in the video application to obtain a third video clip corresponding to the target object, wherein the keyword is obtained based on the target object;

[0163] The search result submodule is used to summarize the first video clip, the second video clip and / or the third video clip to obtain a video search result.

[0164] Furthermore, the first search submodule may include:

[0165] a television channel locking unit, configured to obtain a television channel containing the target object according to a pre-stored television channel list when detecting that the current channel of the television is a television channel;

[0166] The video search unit is used to perform a keyword search according to the parameter list of the television channel to obtain the corresponding first video clip.

[0167] Furthermore, the first search submodule may further include:

[0168] The channel summary table acquisition unit is used to summarize the parameter list of each TV channel to obtain the TV channel summary table, wherein the parameter list includes at least one of a program name list, a program content list and a channel label list.

[0169] Furthermore, the first search submodule may further include:

[0170] a channel label list acquisition unit, configured to, upon detecting that the current channel of the television is a television channel and a television program is being played, capture the television picture of the current television channel according to a preset time interval to obtain a screenshot image;

[0171] Performing image recognition on the screenshot image to obtain all objects in the screenshot image;

[0172] Counting all objects in all the screenshots within a preset time period to obtain label information of the video clips corresponding to the preset time period;

[0173] A channel label list of the current television channel is obtained according to the video clip and the label information.

[0174] Furthermore, the channel label list acquiring unit may further include:

[0175] a custom label unit, configured to display the obtained statistical results of all objects on the display interface of the television when receiving an operation instruction from the user requesting to edit the label, wherein the statistical results include the frequency of occurrence of the objects;

[0176] According to the user's operation instruction of adding or deleting an object, the object is added or deleted in the statistical result accordingly to obtain the edited statistical result;

[0177] According to the preset filtering conditions, the objects in the edited statistical results are filtered to obtain the label information of the video clips corresponding to the preset time period.

[0178] Furthermore, the object includes at least one of an object, a person, and a scene; and the image recognition module may include:

[0179] an object recognition submodule, configured to perform image recognition on the input image using image recognition technology, obtain all objects in the input image, and display them on a display interface of the television;

[0180] The object selection submodule is used to take the object selected by the user as the target object when an operation instruction for selecting an object is obtained from the user.

[0181] It should be noted that the functions and corresponding technical effects achievable by each module in the TV video search device provided in this embodiment can be referred to the description of the specific implementation methods in each embodiment of the TV video search method of the present invention. For the sake of brevity, they will not be repeated here.

[0182] Example 4

[0183] Based on the same invention concept, Figure 2 , which is a schematic diagram of the hardware structure of a television involved in various embodiments of the present invention. This embodiment provides a television that may include a processor and a memory. The memory stores a computer program that, when executed by the processor, implements all or part of the steps of various embodiments of the television video search method of the present invention.

[0184] Specifically, the television refers to a smart television or a television set-top box that can achieve network connection.

[0185] It can be understood that the device may further include a communication bus, a user interface and a network interface.

[0186] The communication bus is used to realize the connection and communication between these components.

[0187] The user interface is used to connect to the client and communicate data with the client. The user interface may include an output unit, such as a display screen, and an input unit, such as a keyboard and a camera. Optionally, the user interface may also include other input / output interfaces, such as a standard wired interface and a wireless interface, such as a wirelessly connected TV remote control.

[0188] The network interface is used to connect to the backend server and perform data communication with the backend server. The network interface may include an input / output interface, such as a standard wired interface, or a wireless interface, such as a Wi-Fi interface.

[0189] The memory is used to store various types of data, which may include, for example, instructions for any application or method in the television, as well as application-related data. The memory may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. Optionally, the memory may also be a storage device independent of the processor.

[0190] The processor is used to call the computer program stored in the memory and execute the television video search method as described above. The processor can be an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor, or other electronic components, and is used to execute all or part of the steps of each embodiment of the television video search method as described above.

[0191] Example 5

[0192] Based on the same inventive concept, this embodiment provides a computer-readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a disk, an optical disk, a server, an App application store, etc. The storage medium stores a computer program, and the computer program can be executed by one or more processors. When the computer program is executed by the processor, it can implement all or part of the steps of each embodiment of the TV video search method of the present invention.

[0193] The above descriptions are merely optional embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by utilizing the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly or indirectly applied in other related technical fields, are included in the patent protection scope of the present invention.

Claims

1. A method for searching television videos, characterized in that: The method comprises: Get the input image; Performing image recognition on the input image to obtain a target object; wherein the object to be image recognized includes at least one of an object, a person, and a scene; According to the current channel of the television, a video clip corresponding to the target object is searched in pre-stored information to obtain a video search result; wherein the pre-stored information includes a TV channel list, local storage content of the video application, and / or content provided by the video application server, the TV channel list includes the content of each TV channel in each time period, the content includes the program name, program content summary, and a custom channel label, the custom channel label is established based on the object information of the TV channel watched by the user in each time period and / or the object information of the TV channel watched by the user in the time period watched by the user; The step of searching for a video clip corresponding to the target object in pre-stored information according to the current channel of the television to obtain a video search result specifically includes: When it is detected that the current channel of the television is a television channel, obtaining a first video clip corresponding to the target object according to a pre-stored parameter list; When it is detected that the current channel of the television is a video application channel, obtaining a second video clip corresponding to the target object according to the local storage content of the video application; When it is detected that the current channel of the television is a video application channel, performing a keyword search in the video application according to the content provided by the video application server to obtain a third video clip corresponding to the target object, wherein the keyword is obtained based on the target object; Summarizing the first video clip, the second video clip, and / or the third video clip to obtain a video search result; The step of obtaining the first video clip corresponding to the target object according to a pre-stored parameter list when the current channel of the television is detected to be a television channel specifically includes: When it is detected that the current channel of the television is a television channel channel, obtaining a target television channel containing the target object according to the pre-stored television channel list; In the target TV channel, a keyword search is performed according to a parameter list of the target TV channel to obtain a corresponding first video clip, wherein the parameter list of the target TV channel includes a parameter list combined with a program name, a program content summary, and a custom channel label; The step of obtaining a target TV channel containing the target object according to the pre-stored TV channel list includes: Search in the TV channel master list to obtain all TV channels containing the target object, and automatically jump to the TV channel containing the target object closest to the current TV channel to obtain the target TV channel.

2. The television video search method according to claim 1, wherein: Before the step of obtaining the TV channel containing the target object according to a pre-stored TV channel list when the current channel of the TV is detected to be a TV channel, the method further includes: The parameter list of each TV channel is summarized to obtain the TV channel summary list, wherein the parameter list of each TV channel includes at least one of a program name list, a program content list and a channel label list, and the parameter list of each TV channel is established and updated based on the TV channels and / or time periods watched by the user.

3. The television video search method according to claim 2, wherein: Before the step of aggregating the parameter lists of each television channel to obtain a total list of television channels, the method further includes: When it is detected that the current channel of the television is a television channel and a television program is being played, the television picture of the current television channel is captured according to a preset time interval to obtain a screenshot image; Performing image recognition on the screenshot image to obtain all objects in the screenshot image; Counting all objects in all the screenshots within a preset time period to obtain label information of the video clips corresponding to the preset time period; A channel label list of the current television channel is obtained according to the video clip and the label information.

4. The television video search method according to claim 3, wherein: After the step of counting all objects in all the screenshots within a preset time period to obtain label information of the video segments corresponding to the preset time period, the method further includes: When an operation instruction requesting to edit a tag is obtained from the user, the obtained statistical results of all objects are displayed on the display interface of the television, wherein the statistical results include the frequency of occurrence of the objects; According to the user's operation instruction of adding or deleting an object, the object is added or deleted in the statistical result accordingly to obtain the edited statistical result; According to the preset filtering conditions, the objects in the edited statistical results are filtered to obtain the label information of the video clips corresponding to the preset time period.

5. The television video search method according to any one of claims 1 to 4, characterized in that: The step of performing image recognition on the input image to obtain the target object specifically includes: Performing image recognition on the input image using image recognition technology to obtain all objects in the input image and displaying them on the display interface of the television; When an operation instruction for selecting an object is obtained from the user, the object selected by the user is used as the target object.

6. A television video search device, characterized in that: The device comprises: An image acquisition module, used to acquire an input image; An image recognition module, configured to perform image recognition on the input image to obtain a target object; wherein the object to be image recognized includes at least one of an object, a person, and a scene; A video search module is configured to search pre-stored information for a video clip corresponding to a target object based on the current channel of the television, thereby obtaining a video search result; wherein the pre-stored information includes a TV channel list, locally stored content of a video application, and / or content provided by a video application server; the TV channel list includes the content of each TV channel in each time period, the content including the program name, a brief description of the program content, and a custom channel label; the custom channel label is established based on the object information of the TV channels watched by the user in each time period and / or the object information of the TV channels watched by the user in the time period watched by the user; The video search module includes: a first search submodule, configured to obtain a first video clip corresponding to the target object according to a pre-stored parameter list when detecting that the current channel of the television is a television channel; A second search submodule is configured to obtain a second video clip corresponding to the target object based on local storage content of the video application when it is detected that the current channel of the television is a video application channel; a third search submodule configured to, when detecting that the current channel of the television is a video application channel, perform a keyword search in the video application based on the content provided by the video application server to obtain a third video clip corresponding to the target object, wherein the keyword is obtained based on the target object; a search result submodule, configured to aggregate the first video clip, the second video clip, and / or the third video clip to obtain a video search result; The first search submodule includes: a television channel locking unit, configured to obtain a target television channel containing the target object according to a pre-stored television channel list when detecting that the current channel of the television is a television channel; a video search unit, configured to perform a keyword search in the target television channel according to a parameter list of the target television channel to obtain a corresponding first video clip, wherein the parameter list of the target television channel includes a parameter list composed of a program name, a program content summary, and a custom channel label; The step of obtaining a target TV channel containing the target object according to the pre-stored TV channel list includes: Search in the TV channel master list to obtain all TV channels containing the target object, and automatically jump to the TV channel containing the target object closest to the current TV channel to obtain the target TV channel.

7. A television, characterized in that: The television includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the television video search method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program can be executed by one or more processors to implement the television video search method according to any one of claims 1 to 5.

Images