Intelligent AI information processing method and system, and intelligent terminal

Abstract

The invention relates to an intelligent AI information processing method and system and an intelligent terminal, and relates to the technical field of information processing, and the method comprises the steps: obtaining sound detection information in a preset detection range and triggering information of a user; inputting the sound detection information into a preset noise model to match noise detection information; updating the sound detection information through a preset sound processing method according to the noise detection information; inputting the updated sound detection information into a preset audio processing system, and determining an output text according to the updated sound detection information when the audio processing system outputs the text; obtaining a selected domain image based on the trigger information; determining a selected text field according to the selected field image; and performing matching combination according to the selected text field and the output text to obtain final uploading information, and uploading the final uploading information to a preset storage database for storage. The information processing method and device have the effect of improving the information processing accuracy.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to an information processing method, system and intelligent terminal for intelligent AI. Background Art

[0002] Information processing refers to the process of collecting, storing, processing, transmitting, retrieving and outputting various forms of information.

[0003] Information includes images, tables, text, and sounds, etc. Information processing is usually performed manually to better utilize information resources and realize the value of information. When a user undergoes a physical examination, the user needs to manually fill in the corresponding information according to different departments, and when the user is undergoing the examination, the doctor also needs to fill in the information based on the results of the user's examination, and then mark, annotate, or match and organize the content filled in by the doctor, the corresponding pictures and tables of the examination results, and the examination items of the corresponding department to facilitate subsequent inquiries.

[0004] When multiple information such as text, images, and sounds need to be processed, manually matching the written content with other information such as images or tables is prone to errors and omissions and repeated operations, resulting in mismatches in the recorded information and reducing the accuracy of information processing. This needs to be improved. Summary of the invention

[0005] In order to improve the accuracy of information processing, the present invention provides an intelligent AI information processing method, system and intelligent terminal.

[0006] In a first aspect, the present invention provides an intelligent AI information processing method, which adopts the following technical solution: An intelligent AI information processing method, comprising: Obtaining sound detection information within a preset detection range and user trigger information; Matching noise detection information based on sound detection information input into a preset noise model; Update the sound detection information by a preset sound processing method according to the noise detection information; Inputting the updated sound detection information into a preset audio processing system, and determining the output text according to the updated sound detection information when the audio processing system outputs text; Acquiring an image of a selected area based on trigger information; Determine the selected text field according to the selected field image; The selected text field is matched and combined with the output text to obtain the final upload information and upload it to the preset storage database for storage.

[0007] By adopting the above technical solution, the sound detection information within the detection range is understood to eliminate the noise detection information to obtain new sound detection information and output text, the selected text field is obtained by triggering information, and the final uploaded information is obtained by matching and combining with the output text, and the final uploaded information is uploaded to the storage database for storage, thereby improving the efficiency of information collection and reducing the probability of errors and repeated operations caused by manual matching of information, thereby improving the accuracy of information processing.

[0008] Optional, preset sound processing methods include: Obtaining image information of people within a preset detection range; Eliminate the noise detection information according to the sound detection information to form the actual sound detection information; Matching reference timbre information from a preset timbre database based on the character image information; Determine whether the timbre in the actual sound detection information is all the reference timbre information; Based on the fact that the timbre in the actual sound detection information is all the reference timbre information, the actual sound detection information is used as the updated sound detection information; Based on the fact that the timbre in the actual sound detection information does not contain all the reference timbre information, the sound corresponding to the reference timbre information is eliminated according to the actual sound detection information, and the remaining actual sound detection information is used as the remaining sound information; Updated sound detection information is determined according to the remaining sound information and the character image information.

[0009] By adopting the above technical scheme, the noise detection information in the sound detection information is eliminated to obtain the actual sound detection information, and the presence of benchmark timbre information in the actual sound detection information is analyzed, and the remaining sound information is obtained based on the analysis result, and the remaining sound information and the character image information are analyzed to obtain the updated sound detection information, thereby reducing the impact of external noise and irrelevant sounds on the conversion of text, thereby improving the accuracy of the text converted by the sound detection information.

[0010] Optionally, the method for determining updated sound detection information according to the remaining sound information includes: Determine the selected character image according to the character image information and the trigger information; According to the actual sound detection information, the marked durations corresponding to different timbres in the remaining sound information are obtained; Determine the volume change value and each change time point of the volume change value within the marked duration according to the remaining sound information and the marked duration; When the preset mouth feature does not appear in the selected person image, an actual distance value between a preset acquisition device and the person in the selected person image is obtained, wherein the acquisition device includes an image acquisition device and a sound acquisition device; Update the actual distance value based on the marking time; Calculate the distance deviation value corresponding to each change time point according to the updated actual distance value; Based on the distance deviation value and the remaining sound information, matching the volume estimated change value and the reference deviation range corresponding to the distance deviation value from a preset volume database; Calculate the difference between the volume change value at the corresponding change time point and the volume estimated change value and use it as the volume deviation value; When the volume deviation value falls within the reference deviation range, extract the remaining sound information that falls within the volume deviation value corresponding to the preset reference deviation range from the remaining sound information to form selected sound information, and use the selected sound information as new sound detection information; The selected timbre is retrieved based on the selected sound information, and the selected timbre is recorded into the timbre database for subsequent matching and retrieval.

[0011] By adopting the above technical solution, the character image information and the trigger information are analyzed to obtain the selected character image, and the actual distance value between the character in the selected character image and the collection device and the remaining sound information are understood to obtain the volume deviation value. The situation where the volume deviation value falls within the reference deviation range is analyzed, and the selected sound information is obtained based on the analysis result, and the selected timbre of the selected sound information is stored and recorded, thereby improving the efficiency of information processing while the intelligent AI continues to learn and adapt.

[0012] Optionally, the method for determining updated sound detection information according to the remaining sound information further includes: When a preset mouth feature appears in the selected person image, an image corresponding to the preset head feature is framed based on the selected person image, and the framed image is used as the head image; Determine the mouth shape corresponding to the mouth features according to the head image; Determine the time point of the mouth change within the marked time according to the mouth shape and the marked time; Determine the estimated sound output time point based on the mouth change time point and mouth shape; Based on the overlap comparison between the estimated sound output time point and the change time point, the residual sound information corresponding to the overlap between the estimated sound output time point and the change time point is extracted from the residual sound information to form the target selected sound, and the target selected sound is used as the updated sound detection information; The target selected sound is retrieved based on the target selection sound, and the target selected sound is recorded into the sound database to facilitate subsequent matching and retrieval.

[0013] By adopting the above technical solution, when the mouth feature appears in the selected character image, the head image and the mouth shape are selected from the selected character image, and the mouth shape is analyzed to obtain the estimated sound output time point, and the overlap comparison is performed with the change time point. Based on the comparison result, the target selected sound is obtained and used as the updated sound detection information. The target selected timbre is entered into the timbre database to facilitate subsequent matching and retrieval, thereby being able to identify the sound required by the user and improve the accuracy of sound collection.

[0014] Optionally, the method for matching and combining the selected domain image with the output text includes: Identify different dividing lines based on the selected domain image, and combine based on the different dividing lines to obtain image modules; Identify color information corresponding to each image module from a preset color database based on the selected field image; Combining image modules of the same color based on color information to form actual area images of different color information; Determine the location of the regional image according to the actual regional image and the selected area image; Based on the regional image position and the actual regional image, an actual regional image corresponding to the regional image position not falling into other actual regional images is selected and defined as a selected regional image, and each actual regional image in the selected regional image is used as a matching regional image; Acquire the trigger range according to the trigger information; Selecting a target area image from the selected area image based on the trigger range; Matching image domain information from a preset image database according to the target area image, the matching area image, and the area image position corresponding to the matching area image; Based on the output text, match and output different text fields from a preset text database; Based on the output text, the text with consistent information in the text field and the image field is extracted for matching and combination.

[0015] By adopting the above technical solution, the boundary lines and color information in the selected area image are identified and matched and combined to obtain the selected area image and the matching area image, and the target area image is obtained through triggering information. Then, the image area information is obtained through the target area image, the matching area image and the area image position corresponding to the matching area image, and matched and combined with the text in the corresponding text area in the output text, thereby improving the accuracy of information processing while recording and processing the information.

[0016] Optionally, the method of selecting the target area image includes: Determine whether there are multiple selected area images within the trigger range; If there are no multiple selected area images within the trigger range, the selected area image corresponding to the trigger range is used as the target area image; If there are multiple selected area images within the trigger range, the total triggered image is determined based on the trigger range and the selected area images; Determine the magnification according to the correspondence between the triggered total image and the selected area image; determining an estimated center position based on the trigger range, and magnifying the selected area image by a magnification factor based on the estimated center position to form a new selected area image; After the selected area image is enlarged, the trigger information is reacquired, and selection is continued from the new selected area image according to the reacquired trigger information until there is only one selected area image in the reacquired trigger information, and the selected area image corresponding to the reacquired trigger range is used as the target area image.

[0017] By adopting the above technical solution, the number corresponding to the selected area images within the trigger range is understood to obtain the total triggered image, and the magnification factor is obtained by triggering the total image and the selected area image and amplifying the selected area image. After the selected area image is amplified, the trigger information is reacquired until there is only one selected area image within the trigger range and it is used as the target area image, thereby improving the accuracy of matching the selected image with the output text.

[0018] Optionally, the method for obtaining the actual distance value includes: Determine whether the actual distance value is less than a preset reference distance value; If the actual distance value is less than the reference distance value, continue to determine the distance deviation value; If the actual distance value is not less than the reference distance value, the preset collection device is controlled to operate based on the reference distance value; Inputting the updated sound detection information into a preset timbre database to match different actual timbre parameters, and using the number of the actual timbre parameters as the actual timbre number; Determine whether the actual number of timbres is one; If the actual number of timbre is one, continue to control the preset acquisition device to follow and shoot at the reference distance value; If the actual number of timbre is not one, the preset acquisition device is controlled to perform follow-up shooting according to the character image information.

[0019] By adopting the above technical solution, the actual distance value is compared with the reference distance value, and the operation of the acquisition device is flexibly adjusted according to the comparison result and the actual number of timbres, so as to follow and detect the person whose information the user needs to obtain, thereby ensuring the integrity of the information required by the user.

[0020] Optionally, the method of controlling a preset acquisition device to perform follow-up shooting according to the character image information includes: When the actual number of timbre is not one, obtaining a reference shooting range of the acquisition device; Select corresponding character image information according to actual timbre parameters and define it as a follow-up character image; Determine the actual following position according to the image of the person being followed and the preset image detection information; Calculate the distance between each actual following position as the actual interval distance; The actual shooting range is obtained by combining different numbers of actual following positions with actual interval distances; Based on different actual shooting ranges, adjacent combinations are performed to form different actual shooting total ranges and shooting center positions corresponding to the actual shooting total ranges; Determine the minimum number of shots that can be covered by the benchmark shooting range based on different actual total shooting ranges and benchmark shooting ranges; Based on the minimum shooting quantity, a corresponding number of acquisition devices are output from the preset reset position, and follow-up shooting is performed based on the shooting center position and the reference distance value.

[0021] By adopting the above technical solution, the actual following position and the actual interval distance are understood to obtain the actual total shooting range and the shooting center position, and the minimum shooting number is obtained by the actual total shooting range and the benchmark shooting range. Then, the acquisition device is controlled to output with the minimum shooting number, and the person corresponding to the actual timbre parameter is followed and shot with the shooting center position and the benchmark distance value, thereby reducing the energy output of the acquisition device while ensuring the integrity of the information required by the user.

[0022] In the second aspect, the present application provides an intelligent AI information processing system, which adopts the following technical solutions: An intelligent AI information processing system, comprising: An acquisition module is used to acquire sound detection information, trigger information, selected area image, person image information, marking duration, actual distance value, trigger range, and reference shooting range; A memory for storing a program of an information processing method of an intelligent AI; The processor is used to load, execute and implement the program stored in the memory.

[0023] In a third aspect, the present application provides a smart terminal, which adopts the following technical solution: An intelligent terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute an intelligent AI information processing method.

[0024] In summary, the present application includes at least one of the following beneficial technical effects: 1. By understanding the sound detection information within the detection range to eliminate the noise detection information to obtain new sound detection information and output text, the selected text field is obtained through triggering information, and the final uploaded information is obtained by matching and combining with the output text, and the final uploaded information is uploaded to the storage database for storage, thereby improving the efficiency of information collection and reducing the probability of errors and repeated operations caused by manual matching of information, thereby improving the accuracy of information processing; 2. Analyze the character image information and trigger information to obtain the selected character image, and understand the actual distance value between the character in the selected character image and the acquisition device and the remaining sound information to obtain the volume deviation value, analyze the volume deviation value falling within the reference deviation range, and obtain the selected sound information based on the analysis results, and store and record the selected timbre of the selected sound information, so as to improve the efficiency of information processing while the intelligent AI continues to learn and adapt; 3. The actual total shooting range and the shooting center position are obtained by understanding the actual following position and the actual interval distance, and the minimum shooting number is obtained by the actual total shooting range and the reference shooting range. Then, the acquisition device is controlled to output with the minimum shooting number, and the person corresponding to the actual timbre parameters is followed and shot with the shooting center position and the reference distance value, thereby reducing the energy output of the acquisition device while ensuring the integrity of the information required by the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a method flow chart of an intelligent AI information processing method according to an embodiment of the present invention; Figure 2 is a method flow chart of a preset sound processing method according to an embodiment of the present invention; Figure 3 The method flow of the embodiment of the present invention is to determine the updated sound detection information according to the remaining sound information Figure 1 ; Figure 4 The method flow of the embodiment of the present invention is to determine the updated sound detection information according to the remaining sound information Figure 2 . DETAILED DESCRIPTION

[0026] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0027] Reference Figure 1 , the embodiment of the present application discloses an intelligent AI information processing method, comprising the following steps: Step S100: Acquire sound detection information within a preset detection range and user trigger information.

[0028] The detection range is the range set by the technicians where images and sounds need to be collected. Sound detection information refers to information such as the volume and timbre of the sound emitted within the detection range. The volume and timbre of the sound within the detection range are collected and combined by a preset collection device to form sound detection information. The collection device refers to a device that can slide on a preset track on the wall and is used to collect images and sounds. The collection device includes an image collection device and a sound collection device. The image collection device can be a camera, and the sound collection device can be a microphone. Trigger information refers to control information for users to output pictures and sounds through a mobile terminal. The information clicked by the user is retrieved from the system of the mobile terminal corresponding to the user as trigger information.

[0029] Step S101: inputting the sound detection information into a preset noise model to match the noise detection information.

[0030] Noise detection information refers to information such as the timbre and type of noise in the surroundings that affects the audio processing system's recognition of text. The noise detection information is input into a preset noise model to match the noise detection information. The noise model is a model trained by the operator inputting the timbre and type of noise emitted by different objects in advance, which will not be described in detail here. The audio processing system refers to the software system used to convert sound into text.

[0031] Step S102: updating the sound detection information through a preset sound processing method according to the noise detection information.

[0032] The sound processing method refers to a method for processing the sound detection information. For specific processing steps, refer to step S200 to step S405. The sound detection information is processed by the sound processing method and the noise detection information to improve the accuracy of the audio processing system in recognizing text.

[0033] Step S103: inputting the updated sound detection information into a preset audio processing system, and determining the output text according to the updated sound detection information when the audio processing system outputs text.

[0034] The output text refers to the text information expressed by the sound detection information. By inputting the updated sound detection information into a preset audio processing system, when the audio processing system outputs text, it indicates that the sound detection information is information transmission for a person, so the updated sound detection information is converted into text through the audio processing system, and the converted text is used as the output text.

[0035] Step S104: Acquire a selected area image based on the trigger information.

[0036] The selected area image refers to an image captured when a user outputs trigger information through a mobile terminal, and the image captured by the trigger information is retrieved from the system of the mobile terminal as the selected area image.

[0037] Step S105: determining the selected text area according to the selected area image.

[0038] The selected text area refers to the text area corresponding to the selected area image. The selected text area is matched from a preset image database through the selected area image. The image database stores text areas expressed by different images. The image database is a manually set database and will not be described in detail here.

[0039] Step S106: matching and combining the selected text field with the output text to obtain final upload information and uploading it to a preset storage database for storage.

[0040] The final uploaded information refers to the information that is ultimately available for retrieval and viewing by users or relevant personnel. The information generated by marking or annotating the image corresponding to the field corresponding to the output text and the selected text field is used as the final uploaded information, and the final uploaded information is uploaded to a preset storage database for storage to facilitate retrieval and viewing by users or relevant personnel, thereby improving the efficiency of information collection and reducing the probability of errors and repeated operations caused by manual matching of information, thereby improving the accuracy of information processing.

[0041] Reference Figure 2 , the preset sound processing methods include: Step S200: Acquire image information of a person within a preset detection range.

[0042] The character image information refers to the image of the corresponding character within the detection range. The image within the detection range is captured by the acquisition device, and the image corresponding to the preset character features is selected as the character image information. Character features refer to the shape and color and other features corresponding to each posture of the character. The character features are set in advance by those skilled in the art and will not be described in detail here.

[0043] Step S201: Eliminate the noise detection information according to the sound detection information to form the actual sound detection information.

[0044] The actual sound detection information refers to the sound detection information corresponding to the noise processing, and the noise detection information in the sound detection information is eliminated, and the sound detection information after the elimination is used as the actual sound detection information.

[0045] Step S202: matching reference timbre information from a preset timbre database based on the character image information.

[0046] The reference timbre information refers to the reference timbre of the character corresponding to the character image information, and the reference timbre information is matched from a preset timbre database through the character image information. The timbre database stores timbre corresponding to different characters, and the timbre database is a database set by humans, which will not be described in detail here.

[0047] Step S203: Determine whether the timbre in the actual sound detection information is all reference timbre information.

[0048] By determining whether the timbre in the actual sound detection information is all reference timbre information, it is determined whether there are relevant personnel who are not entered into the timbre database to provide explanations.

[0049] Step S204: Based on the fact that the timbre in the actual sound detection information is all the reference timbre information, the actual sound detection information is used as the updated sound detection information.

[0050] When the timbre in the actual sound detection information is all reference timbre information, it means that there is no relevant person who has not entered the timbre database to explain, so the actual sound detection information is used as the updated sound detection information.

[0051] Step S205: based on the fact that the timbre in the actual sound detection information does not all correspond to the reference timbre information, the sound corresponding to the reference timbre information is eliminated according to the actual sound detection information, and the remaining actual sound detection information is used as the remaining sound information.

[0052] When the timbre in the actual sound detection information does not contain all the reference timbre information, it means that there are relevant personnel who have not been entered into the timbre database for explanation. Therefore, the sound corresponding to the reference timbre information in the actual sound detection information is eliminated, and the remaining actual sound detection information is used as the remaining sound information.

[0053] Step S206: determining updated sound detection information according to the remaining sound information and the person image information.

[0054] New sound detection information is obtained by analyzing the remaining sound information and the person image information. The specific analysis steps refer to steps S300 to S405, so that the sound emitted by the relevant person whose voice is not recorded in advance can be recorded and recognized.

[0055] Reference Figure 3 The method for determining updated sound detection information according to the remaining sound information includes: Step S300: Determine a selected character image according to the character image information and the trigger information.

[0056] The selected character image refers to the character image information selected by the user. The character image information within the detection range is displayed on the user's mobile terminal. The character image information at the position corresponding to the trigger information is retrieved from the system as the selected character image.

[0057] Step S301: Acquire the marked durations corresponding to different timbres in the remaining sound information according to the actual sound detection information.

[0058] The marked duration refers to the time length value of the appearance of different timbres in the remaining sound information. The time points at which each timbre appears in the actual sound detection information are marked, and the time points of different timbres in the remaining sound information are retrieved from the marked time points and combined to form a time length value as the marked duration.

[0059] Step S302: determining the volume change value and each change time point of the volume change value within the marked time according to the remaining sound information and the marked time.

[0060] The volume change value refers to the volume change value corresponding to the remaining sound information within the marked duration. The volume of the corresponding single timbre in the remaining sound information at the beginning of the marked duration is taken as the reference volume, and the other volumes of the corresponding single timbre in the remaining sound information within the remaining marked duration are taken as the remaining volumes. The difference between the reference volume and the remaining volumes is calculated as the volume change value. The change time point refers to the time point corresponding to the volume change value of the remaining sound information within the marked duration. The time point corresponding to the volume change value of the single timbre is taken from the marked duration as the change time point.

[0061] Step S303: when the preset mouth feature does not appear in the selected person image, the actual distance value between the preset acquisition device and the person in the selected person image is obtained, and the acquisition device includes an image acquisition device and a sound acquisition device.

[0062] Mouth features refer to the shape, color, and position of a person's mouth. Mouth features are set in advance by a person skilled in the art and will not be described in detail here. The actual distance value refers to the actual distance value between the acquisition device and the person in the selected person image. When the selected person image does not have a mouth feature, it means that it is not easy for the system to know whether the person in the selected person image makes a sound. Therefore, the distance parameter detected by the infrared sensor preset on the acquisition device while the acquisition device continuously captures the selected person image is used as the actual distance value.

[0063] Step S304: updating the actual distance value based on the marking duration.

[0064] By retrieving the distance value within the marked duration from all the changed actual distance values ​​as the new actual distance value, subsequent analysis is facilitated.

[0065] Step S305: Calculate the distance deviation value corresponding to each change time point according to the updated actual distance value.

[0066] The distance deviation value refers to the change in the distance corresponding to the updated actual distance value within the marking duration. The actual distance value at the beginning of the marking duration is taken as the baseline value, and the actual distance values ​​within the remaining marking duration are taken as the remaining values. The difference between the baseline value and the remaining values ​​is calculated as the distance deviation value.

[0067] Step S306: matching the volume estimation change value and the reference deviation range corresponding to the distance deviation value from a preset volume database based on the distance deviation value and the remaining sound information.

[0068] The estimated volume change value refers to the estimated volume change value generated when the actual distance value changes, and the estimated volume change value is matched from a preset volume database through the distance deviation value and the remaining sound information. The reference deviation range refers to the reference deviation range corresponding to the estimated volume change value generated when the actual distance value changes, and the reference deviation range is matched from the volume database through the distance deviation value. The volume database stores different volumes in the remaining sound information and estimated volume change values ​​corresponding to the distance deviation values, and the volume database also stores reference deviation ranges corresponding to different distance deviation values. The volume database is a manually set database and will not be described in detail here.

[0069] Step S307: Calculate the difference between the volume change value at the corresponding change time point and the volume estimated change value and use it as the volume deviation value.

[0070] The volume deviation value refers to the deviation value between the volume change value at the same change time point and the volume estimated change value. The volume change value and the volume estimated change value at the same change time point are retrieved, the difference between the two is calculated, and the difference is used as the volume deviation value.

[0071] Step S308: When the volume deviation value falls within the reference deviation range, extract the remaining sound information that falls within the volume deviation value corresponding to the preset reference deviation range from the remaining sound information to form selected sound information, and use the selected sound information as new sound detection information.

[0072] The selected sound information refers to the sound information emitted by the person corresponding to the selected person image. When the volume deviation value falls into the benchmark deviation range, it means that the sound information emitted by the person corresponding to the selected person image exists in the remaining sound information. Therefore, the remaining sound information corresponding to the volume deviation value falling into the preset benchmark deviation range is extracted from the remaining sound information to form the selected sound information, and the selected sound information is used as the new sound detection information.

[0073] Step S309: Retrieving a selected timbre based on the selected sound information, and recording the selected timbre into a timbre database for subsequent matching and retrieval.

[0074] Selecting a timbre refers to selecting the timbre of the voice of the person corresponding to the selected person image, by retrieving the selected timbre from the selected voice information and recording the selected timbre into the timbre database for subsequent matching and retrieval. This enables the user to identify the voice required by the user and improves the accuracy of voice collection.

[0075] Reference Figure 4 The method for determining updated sound detection information according to the remaining sound information further includes: Step S400: When a preset mouth feature appears in the selected person image, an image corresponding to the preset head feature is framed based on the selected person image, and the framed image is used as the head image.

[0076] The head features refer to the shape, color distribution and position of the head corresponding to the character. The head features are set in advance by those skilled in the art and will not be described in detail here. The head image refers to the head image of the character corresponding to the selected character image. When the mouth feature appears in the selected character image, it means that the system can know whether the sound is made by the character in the selected character image through the mouth feature, so the image corresponding to the head feature is selected from the selected character image frame as the head image.

[0077] Step S401: determining the mouth shape corresponding to the mouth features according to the head image.

[0078] The mouth shape refers to the shape of the mouth of the person corresponding to the selected person image, and the shape corresponding to the mouth feature is selected from the head image as the mouth shape.

[0079] Step S402: determining the mouth change time point within the marking time according to the mouth shape and the marking time.

[0080] The mouth change time point refers to the time point at which the mouth shape changes within the marked duration. The mouth shape at the beginning of the marked duration is taken as the reference shape, and the mouth shape within the remaining marked duration is taken as the remaining shape. The time point when the remaining shape is inconsistent with the reference shape is taken as the mouth change time point.

[0081] Step S403: Determine the estimated sound output time point according to the mouth change time point and the mouth shape.

[0082] The estimated sound output time point refers to the estimated time point when the mouth shape corresponds to the speech. The mouth shape is matched from the preset mouth model to the shape of the mouth when it is open, and then the mouth change time point corresponding to the shape of the mouth when it is open is retrieved from each mouth shape as the estimated sound output time point. The mouth model is a model that technicians input the mouth shape in advance for training to output the shape corresponding to the mouth when it is open, which will not be described in detail here.

[0083] Step S404: Based on the overlap comparison between the estimated sound output time point and the change time point, the remaining sound information corresponding to the overlap between the estimated sound output time point and the change time point is extracted from the remaining sound information to form a target selected sound, and the target selected sound is used as the updated sound detection information.

[0084] Target selected sound refers to identifying the sound information emitted by the person corresponding to the selected person image by recognizing the mouth shape, and extracting the remaining sound information corresponding to the coincidence of the estimated sound output time point and the change time point from the remaining sound information to form the target selected sound, and using the target selected sound as the updated sound detection information, so as to identify the sound required by the user and improve the accuracy of sound collection.

[0085] Step S405: Retrieving the target selected timbre based on the target selected sound, and recording the target selected timbre into the timbre database for subsequent matching and retrieval.

[0086] The target selected timbre refers to the timbre of the sound corresponding to the target selected sound. The target selected timbre is retrieved from the target selected sound and entered into the timbre database for facilitating subsequent matching and retrieval, thereby facilitating subsequent direct retrieval of the timbre for sound analysis, thereby improving the efficiency of information processing.

[0087] Methods for matching and combining images with output text according to selected domains include: Step S500: identifying different boundary lines based on the selected field image, and combining the different boundary lines to obtain an image module.

[0088] The demarcation lines refer to the lines in the selected domain image, and the lines identified from the selected domain image are used as demarcation lines. The image modules refer to the modules corresponding to the different images formed by the demarcation lines, and the modules corresponding to the images formed by combining the adjacent and abutting demarcation lines are used as image modules.

[0089] Step S501: identifying color information corresponding to each image module from a preset color database based on a selected domain image.

[0090] Color information refers to the color corresponding to each image module. Different colors are identified from a preset color database by selecting a domain image, and each image module is matched and divided corresponding to each color in the selected domain image to form color information corresponding to each image module. The color database stores various colors that can be identified by different images. The color database is a database set by humans and will not be described in detail here.

[0091] Step S502: combining image modules of the same color based on color information to form actual area images of different color information.

[0092] The actual area image refers to the image corresponding to the area with the same color information, and is obtained by combining adjacent image modules with the same color information to obtain actual area images with different color information. For example, there are a table, a mouse pad and a mouse in the selected area image, and the colors of the table, the mouse pad and the mouse are different, and the dividing lines of the table, the mouse pad and the mouse are generally non-intersecting. There are multiple dividing lines on the mouse, but the image modules on the mouse are adjacent and have the same color, so the image of the mouse is an actual area image.

[0093] Step S503: determining the position of the regional image according to the actual regional image and the selected region image.

[0094] The regional image position refers to the position of the actual regional image in the selected domain image, and the position corresponding to the actual regional image is selected in the selected domain image as the regional image position.

[0095] Step S504: based on the region image position and the actual region image, select the actual region image corresponding to the region image position not falling into other actual region images as the selected region image, and use each actual region image in the selected region image as the matching region image.

[0096] The selected area image refers to the actual area image corresponding to the area image position that does not fall into other actual area images, and the actual area image corresponding to the area image position that does not fall into other actual area images is selected from various area image positions and actual area images as the selected area image. The matching area image refers to the actual area image used for retrieval in conjunction with the selected area image, and each actual area image in the selected area image is used as the matching area image. For example, the selected area image is a mouse, and the mouse includes marks, a scroll wheel, and buttons of different colors, so the marks, the scroll wheel, and the buttons are the matching area images of the mouse.

[0097] Step S505: Acquire a trigger range according to the trigger information.

[0098] The trigger range refers to the range of operations such as clicks performed by users on the mobile terminal, and the range corresponding to the trigger information is retrieved from the system as the trigger range.

[0099] Step S506: selecting a target area image from the selected area images based on the trigger range.

[0100] The target area image refers to an image selected by the user, and the selected area image within the trigger range is selected from the selected area image as the target area image.

[0101] Step S507: matching image area information from a preset image database according to the target area image, the matching area image and the area image position corresponding to the matching area image.

[0102] Image domain information refers to the text domain information corresponding to the target area image, and the image domain information is matched from a preset image database through the target area image, the matching area image, and the area image position corresponding to the matching area image. The image database stores the correspondence between the target area image, the matching area image, the area image position corresponding to the matching area image, and the image domain information. The image database is a database set by humans and will not be described in detail here. For example, the buttons and the scroll wheel in the mouse are usually set at fixed positions on the mouse, so the image including the buttons and the scroll wheel is known as a mouse through the images of the buttons and the scroll wheel and the position of the image within the outline corresponding to the image of the mouse.

[0103] Step S508: Match and output different text fields from a preset text database based on the output text.

[0104] The text domain refers to the domain of the text expressed by the output text. The output text is input into a preset text database to match different text domains. The text database stores text domains corresponding to different output texts. The text database is a database set by humans and will not be described in detail here.

[0105] Step S509: extracting texts with consistent text domain information and image domain information based on the output text for matching and combination.

[0106] By extracting text with consistent information in the text field and the image field from the output text, and annotating, remarking, and filling in the extracted text and the target area image, when the user undergoes a physical examination and the target area image is a physical examination form, the information in the output text corresponding to the detection range can be extracted to fill in the physical examination form, thereby improving the accuracy of information processing while recording and processing the information.

[0107] Methods for selecting the target area image include: Step S600: Determine whether there are multiple selected area images in the trigger range.

[0108] By determining whether there are multiple selected area images within the trigger range, it is determined whether the user's selection is accurate.

[0109] Step S601: If there are no multiple selected area images in the trigger range, the selected area image corresponding to the trigger range is used as the target area image.

[0110] When there are no multiple selected area images in the trigger range, it means that the user's selection is accurate, so the selected area image corresponding to the trigger range is used as the target area image.

[0111] Step S602: If there are multiple selected area images in the trigger range, determine the total trigger image according to the trigger range and the selected area images.

[0112] The triggered total image refers to all the selected area images corresponding to the trigger range. When there are multiple selected area images in the trigger range, it means that the user has clicked a large range, resulting in inaccurate user selection. Therefore, the selected area images in the trigger range are used as the triggered total image. In this embodiment, when a part of the selected area image appears in the trigger range, the overall selected area image corresponding to the part is added to the triggered total image.

[0113] Step S603: Determine the magnification factor according to the corresponding relationship between the triggered total image and the selected area image.

[0114] The magnification factor refers to the magnification factor of the selected field image, and the magnification factor is matched from a preset image database by triggering the total image and the selected field image. The image database stores the magnification factor corresponding to the trigger total image that can be displayed as a whole in the selected field image. The image database is a manually set database and will not be described in detail here.

[0115] Step S604: determining an estimated center position based on the trigger range, and magnifying the selected area image by a magnification factor based on the estimated center position to form a new selected area image.

[0116] The estimated center position refers to the estimated center position corresponding to the trigger range. The smallest circle that can contain the trigger range is selected by boxing, and the center point of the circle is used as the estimated center position. The selected area image is controlled to be enlarged with the magnification and the estimated center position, and the enlarged image is used as the new selected area image.

[0117] Step S605: After the selected area image is enlarged, the trigger information is reacquired, and selection is continued from the new selected area image according to the reacquired trigger information until there is only one selected area image in the reacquired trigger information, and the selected area image corresponding to the reacquired trigger range is used as the target area image.

[0118] When the selected area image is enlarged, it is more convenient for users to click and select, so the trigger information is re-acquired, and selection continues from the new selected area image according to the re-acquired trigger information until there is only one selected area image in the re-acquired trigger information, and the selected area image corresponding to the re-acquired trigger range is used as the target area image, thereby improving the accuracy of matching the selected image with the output text.

[0119] The methods to obtain the actual distance value include: Step S700: Determine whether the actual distance value is less than a preset reference distance value.

[0120] The reference distance value refers to the maximum distance value at which the sound collection device can collect sound. The reference distance value is set in advance by those skilled in the art and will not be described in detail here. By determining whether the actual distance value is less than the reference distance value, it is determined whether the fixed collection device can continue to collect sound from the moving person.

[0121] Step S701: If the actual distance value is less than the reference distance value, continue to determine the distance deviation value.

[0122] When the actual distance value is less than the reference distance value, it means that the fixed collection device can continue to collect the sound of the moving person, so step S305 is continued.

[0123] Step S702: If the actual distance value is not less than the reference distance value, the preset collection device is controlled to operate based on the reference distance value.

[0124] When the actual distance value is not less than the reference distance value, it means that the person in the person image information has moved beyond the distance where the fixed collection device can no longer continue to collect sound. Therefore, the collection device is controlled to slide on the wall to keep the distance between the collection device and the person within the reference distance value.

[0125] Step S703: inputting the updated sound detection information into a preset timbre database to match different actual timbre parameters, and using the number of the actual timbre parameters as the actual timbre number.

[0126] The actual timbre parameters refer to the actual timbre parameters existing in the updated sound detection information, and the actual timbre quantity refers to the number of actual timbre parameters existing in the updated sound detection information. The updated sound detection information is input into a preset timbre database to match different actual timbre parameters, and the number corresponding to the matched actual timbre parameters is used as the actual timbre quantity.

[0127] Step S704: Determine whether the actual number of timbre is one.

[0128] By determining whether the actual number of timbres is one, it is determined whether it is necessary to use other acquisition devices for follow-up shooting.

[0129] Step S705: If the actual number of timbre is one, continue to control the preset acquisition device to perform follow-up shooting at the reference distance value.

[0130] When the actual number of timbre is one, it means that only one relevant person is speaking, and there is no need to use other acquisition devices for follow-up shooting, so the preset acquisition device continues to be controlled to follow-up shooting with the reference distance box.

[0131] Step S706: If the actual number of timbre is not one, the preset acquisition device is controlled to perform follow-up shooting according to the character image information.

[0132] When the actual number of timbres is not one, it means that there are multiple relevant persons speaking and other acquisition devices are needed for follow-up shooting. Therefore, the character image information is analyzed to control the preset acquisition device to perform follow-up shooting. For the specific analysis steps, refer to steps S800 to S807.

[0133] The method for controlling a preset acquisition device to perform follow-up shooting according to character image information includes: Step S800: When the actual number of timbre is not one, obtaining a reference shooting range of the acquisition device.

[0134] The reference shooting range refers to the reference range of the shooting image corresponding to the acquisition device. When the actual number of timbre is not one, the operator inputs it into the system in advance and retrieves it to obtain the reference shooting range.

[0135] Step S801: Select corresponding character image information according to actual timbre parameters and define it as a follow-up character image.

[0136] The following person image refers to the person image information corresponding to the person that needs to be followed and photographed by the acquisition device. The person to be followed is obtained through actual timbre parameters, and the person image information corresponding to the person is used as the following person image.

[0137] Step S802: determining the actual following position according to the image of the person being followed and preset image detection information.

[0138] The image detection information is the image information corresponding to the detection range taken in advance by the technician. The actual following position refers to the position of the person that needs to be followed by the acquisition device within the detection range, and the position corresponding to the image of the person being followed in the image detection information is used as the actual following position.

[0139] Step S803: Calculate the distance between each actual following position as the actual interval distance.

[0140] The actual interval distance refers to the distance between each actual following position, and the actual interval distance is calculated by calculating the straight line distance between each actual following position. In this embodiment, when the number of actual following positions exceeds three, the straight lines of the actual interval distances corresponding to each actual following position intersect.

[0141] Step S804: combining different numbers of actual following positions with actual interval distances to obtain an actual shooting range.

[0142] The actual shooting range refers to the maximum range formed by connecting the actual following position and the actual interval distance. The maximum range formed by combining the actual following positions corresponding to the reference number and the straight lines corresponding to the actual interval distance is taken as the actual shooting range. When the number of actual following positions exceeds three, the straight lines corresponding to the actual interval distances of the actual following positions intersect, so the range formed by the boundary straight lines corresponding to the actual following position and the actual interval distance is taken as the actual shooting range.

[0143] When the number of actual following positions is two, since the person image information has different sizes, the range between the person image information corresponding to the two actual following positions is taken as the actual shooting range.

[0144] Step S805: performing adjacent combinations based on different actual shooting ranges to form different actual shooting total ranges and shooting center positions corresponding to the actual shooting total ranges.

[0145] The actual total shooting range refers to the total range corresponding to the combination of multiple adjacent actual shooting ranges. Different actual total shooting ranges are formed by combining different actual shooting ranges. The shooting center position refers to the estimated center position corresponding to the actual total shooting range. The smallest circle that can contain the actual total shooting range is obtained through the actual total shooting range, and the center of the circle is used as the shooting center position.

[0146] Step S806: determining the minimum number of shots that can be covered by the reference shooting range according to different actual total shooting ranges and the reference shooting range.

[0147] The minimum shooting number is the minimum number of follow-up shootings that need to be performed using the acquisition device, and the minimum shooting number is obtained by combining and analyzing different actual shooting total ranges and reference shooting ranges. For specific analysis steps, refer to steps S8061 to S8068.

[0148] Step S807: Outputting a corresponding number of acquisition devices from a preset reset position based on the minimum shooting number, and performing follow-up shooting at the shooting center position and the reference distance value.

[0149] The reset position refers to the position where the acquisition device is reset. The reset position is set in advance by a person skilled in the art and will not be described in detail here. A plurality of acquisition devices are arranged in the reset position. By controlling the acquisition device to output from the reset position with the minimum number of shots and follow-shooting with the shooting center position and the reference distance value, the sound and image of all relevant persons within the detection range can be collected.

[0150] Further, the method for determining the minimum number of shots includes: Step S8061: determining different actual shooting shapes and actual shooting areas corresponding to the actual shooting shapes according to different actual shooting total ranges.

[0151] The actual shooting shape refers to the shape corresponding to the actual shooting total range, which is obtained by retrieving the boundary lines of the actual shooting total range and combining the retrieved boundary lines to form a shape as the actual shooting shape. The actual shooting area refers to the area corresponding to the actual shooting shape, which is obtained by retrieving the distance values ​​corresponding to the actual shooting shape from different actual interval distances and combining the retrieved distance values ​​to obtain the actual shooting area.

[0152] Step S8062: Retrieve a reference shooting area and a reference shooting shape based on the reference shooting range.

[0153] The reference shooting shape refers to the shape corresponding to the reference shooting range, and the reference shooting area refers to the area corresponding to the reference shooting shape. The reference shooting area and the reference shooting shape are retrieved from the reference shooting range.

[0154] Step S8063: Based on different actual shooting areas, an actual shooting shape smaller than the reference shooting area is selected as the target shooting shape.

[0155] The target shooting shape refers to an actual shooting shape whose actual shooting area is smaller than the reference shooting area, and an actual shooting shape whose actual shooting area is smaller than the reference shooting area is selected from different actual shooting areas as the target shooting shape. In this embodiment, there are multiple target shooting shapes, and the largest actual shooting area needs to be divided into actual shooting shapes of different areas and non-overlapping ranges as the target shooting shape.

[0156] Step S8064: based on different target shooting shapes, selecting a target shooting shape that falls within the reference shooting shape as a marked shooting shape.

[0157] The marked shooting shape refers to a shape that can be shot by the acquisition device, and the target shooting shape corresponding to the reference shooting shape is selected from different target shooting shapes as the marked shooting shape. There are multiple marked shooting shapes.

[0158] Step S8065: Determine the maximum shooting shape according to different actual total shooting ranges.

[0159] The maximum photographing shape refers to the maximum shape connected by all actual following positions as the maximum photographing shape, and the shape with the largest area corresponding to the actual photographing total range where all actual following positions exist is taken as the maximum photographing shape.

[0160] Step S8066: Determine the remaining shooting shapes according to the marked shooting shape and the maximum shooting shape.

[0161] The remaining captured shape refers to the shape remaining after the marker captured shape is eliminated from the maximum captured shape. The shape formed by eliminating each marker captured shape from the maximum captured shape is the remaining captured shape.

[0162] Step S8067: determining the remaining number of shots according to the different remaining shooting shapes, the reference shooting areas and the reference shooting shapes, and arranging the remaining number of shots in reverse order to obtain the minimum remaining number of shots as the selected number of shots.

[0163] The remaining number of shots refers to the number of shots of the remaining shooting shapes by the acquisition device. The remaining number of shots is obtained by analyzing different remaining shooting shapes, reference shooting areas, and reference shooting shapes. There are multiple remaining shooting numbers. For example, when the marked shooting shape is the shape of the center part corresponding to the maximum shooting shape, the shapes on both sides of the marked shooting shape are the remaining shooting shapes. The area required is smaller than the reference shooting area and the remaining shooting shapes corresponding to the reference shooting shape. Therefore, at least two acquisition devices are required to be able to shoot. The selected shooting number refers to the minimum number of acquisition devices actually required. The selected shooting number is obtained by arranging the different remaining shooting numbers in reverse order to obtain the minimum remaining shooting number.

[0164] Step S8068: Determine the minimum number of shots based on the selected number of shots and the number corresponding to the marked shot shape.

[0165] The minimum shooting number refers to the minimum number of acquisition devices corresponding to the shooting and following. The sum of the selected shooting number and the number of acquisition devices corresponding to the marked shooting shape is calculated, and the sum is used as the minimum shooting number. The number of acquisition devices corresponding to the marked shooting shape is one.

[0166] Based on the same inventive concept, an embodiment of the present invention provides an intelligent AI information processing system, including: The acquisition module is used to obtain sound detection information, trigger information, selected area image, person image information, marking duration, actual distance value, trigger range and reference shooting range.

[0167] A memory for storing a program of an information processing method of an intelligent AI.

[0168] The processor is used to load, execute and implement the program stored in the memory.

[0169] Based on the same inventive concept, an embodiment of the present invention provides an intelligent terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute an intelligent AI information processing method.

[0170] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0171] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An information processing method of intelligent AI, characterized in that: include: Obtaining sound detection information within a preset detection range and user trigger information; Matching noise detection information based on sound detection information input into a preset noise model; Update the sound detection information by a preset sound processing method according to the noise detection information; Inputting the updated sound detection information into a preset audio processing system, and determining the output text according to the updated sound detection information when the audio processing system outputs text; Acquiring an image of a selected area based on trigger information; Determine the selected text field according to the selected field image; Matching and combining the selected text field with the output text to obtain the final upload information and uploading it to a preset storage database for storage; The preset sound processing methods include: Obtaining image information of people within a preset detection range; Eliminate the noise detection information according to the sound detection information to form the actual sound detection information; Matching reference timbre information from a preset timbre database based on the character image information; Determine whether the timbre in the actual sound detection information is all the reference timbre information; Based on the fact that the timbre in the actual sound detection information is all the reference timbre information, the actual sound detection information is used as the updated sound detection information; Based on the fact that the timbre in the actual sound detection information does not contain all the reference timbre information, the sound corresponding to the reference timbre information is eliminated according to the actual sound detection information, and the remaining actual sound detection information is used as the remaining sound information; Determine updated sound detection information according to the remaining sound information and the character image information; The method for determining updated sound detection information according to the remaining sound information includes: Determine the selected character image according to the character image information and the trigger information; According to the actual sound detection information, the marked durations corresponding to different timbres in the remaining sound information are obtained; Determine the volume change value and each change time point of the volume change value within the marked duration according to the remaining sound information and the marked duration; When the preset mouth feature does not appear in the selected person image, an actual distance value between a preset acquisition device and the person in the selected person image is obtained, wherein the acquisition device includes an image acquisition device and a sound acquisition device; Update the actual distance value based on the marking time; Calculate the distance deviation value corresponding to each change time point according to the updated actual distance value; Based on the distance deviation value and the remaining sound information, matching the volume estimated change value and the reference deviation range corresponding to the distance deviation value from a preset volume database; Calculate the difference between the volume change value at the corresponding change time point and the volume estimated change value and use it as the volume deviation value; When the volume deviation value falls within the reference deviation range, extracting the remaining sound information that falls within the volume deviation value corresponding to the preset reference deviation range from the remaining sound information to form selected sound information, and using the selected sound information as new sound detection information; The selected timbre is retrieved based on the selected sound information, and the selected timbre is recorded into the timbre database for subsequent matching and retrieval.

2. The information processing method of an intelligent AI according to claim 1, characterized in that: The method for determining updated sound detection information according to the remaining sound information further includes: When a preset mouth feature appears in the selected person image, an image corresponding to the preset head feature is framed based on the selected person image, and the framed image is used as the head image; Determine the mouth shape corresponding to the mouth features according to the head image; Determine the time point of the mouth change within the marked time according to the mouth shape and the marked time; Determine the estimated sound output time point based on the mouth change time point and mouth shape; Based on the overlap comparison between the estimated sound output time point and the change time point, the remaining sound information corresponding to the overlap between the estimated sound output time point and the change time point is extracted from the remaining sound information to form a target selected sound, and the target selected sound is used as the updated sound detection information; The target selected sound is retrieved based on the target selection sound, and the target selected sound is recorded into the sound database to facilitate subsequent matching and retrieval.

3. The information processing method of an intelligent AI according to claim 1, characterized in that: Methods for matching and combining images with output text according to selected domains include: Identify different dividing lines based on the selected domain image, and combine based on the different dividing lines to obtain image modules; Identify color information corresponding to each image module from a preset color database based on the selected field image; Combining image modules of the same color based on color information to form actual area images of different color information; Determine the location of the regional image according to the actual regional image and the selected area image; Based on the regional image position and the actual regional image, an actual regional image corresponding to the regional image position not falling into other actual regional images is selected and defined as a selected regional image, and each actual regional image in the selected regional image is used as a matching regional image; Acquire the trigger range according to the trigger information; Selecting a target area image from the selected area image based on the trigger range; Matching image domain information from a preset image database according to the target area image, the matching area image, and the area image position corresponding to the matching area image; Based on the output text, match and output different text fields from a preset text database; Based on the output text, the text with consistent information in the text field and the image field is extracted for matching and combination.

4. The information processing method of intelligent AI according to claim 3, characterized in that: Methods for selecting the target area image include: Determine whether there are multiple selected area images within the trigger range; If there are no multiple selected area images within the trigger range, the selected area image corresponding to the trigger range is used as the target area image; If there are multiple selected area images within the trigger range, the total triggered image is determined based on the trigger range and the selected area images; Determine the magnification according to the correspondence between the triggered total image and the selected area image; determining an estimated center position based on the trigger range, and magnifying the selected area image by a magnification factor based on the estimated center position to form a new selected area image; After the selected area image is enlarged, the trigger information is reacquired, and selection is continued from the new selected area image according to the reacquired trigger information until there is only one selected area image in the reacquired trigger information, and the selected area image corresponding to the reacquired trigger range is used as the target area image.

5. The information processing method of intelligent AI according to claim 1, characterized in that: The methods to obtain the actual distance value include: Determine whether the actual distance value is less than a preset reference distance value; If the actual distance value is less than the reference distance value, continue to determine the distance deviation value; If the actual distance value is not less than the reference distance value, the preset collection device is controlled to operate based on the reference distance value; Inputting the updated sound detection information into a preset timbre database to match different actual timbre parameters, and using the number of the actual timbre parameters as the actual timbre number; Determine whether the actual number of timbres is one; If the actual number of timbre is one, continue to control the preset acquisition device to follow and shoot at the reference distance value; If the actual number of timbre is not one, the preset acquisition device is controlled to perform follow-up shooting according to the character image information.

6. The information processing method of intelligent AI according to claim 5, characterized in that: The method for controlling a preset acquisition device to perform follow-up shooting according to character image information includes: When the actual number of timbre is not one, obtaining a reference shooting range of the acquisition device; Select corresponding character image information according to actual timbre parameters and define it as a follow-up character image; Determine the actual following position according to the image of the person being followed and the preset image detection information; Calculate the distance between each actual following position as the actual interval distance; The actual shooting range is obtained by combining different numbers of actual following positions with actual interval distances; Based on different actual shooting ranges, adjacent combinations are performed to form different actual shooting total ranges and shooting center positions corresponding to the actual shooting total ranges; Determine the minimum number of shots that can be covered by the benchmark shooting range based on different actual total shooting ranges and benchmark shooting ranges; Based on the minimum shooting quantity, a corresponding number of acquisition devices are output from the preset reset position, and follow-up shooting is performed based on the shooting center position and the reference distance value.

7. An intelligent AI information processing system, characterized in that: include: An acquisition module is used to acquire sound detection information, trigger information, selected area image, person image information, marking duration, actual distance value, trigger range, and reference shooting range; A memory for storing a program of an intelligent AI information processing method according to any one of claims 1 to 6; The processor is used to load, execute and implement the program stored in the memory.

8. An intelligent terminal, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a computer program which can be loaded by the processor and executes an information processing method of an intelligent AI as claimed in any one of claims 1 to 6.

Images