A film and television LOGO automatic generation method, system, terminal and storage medium
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN COOCAA NETWORK TECH CO LTD
- Filing Date
- 2026-06-08
- Publication Date
- 2026-08-07
AI Technical Summary
[0010]本发明要解决的技术问题在于，针对现有技术缺陷，本发明提供一种影视LOGO自动化生成方法、系统、终端及存储介质，以解决现有技术还存在生成的LOGO文字识别率低、风格与影视内容不匹配以及自动化程度低的问题
本发明通过融合语言理解与图像生成能力，结合自动化审核与后期优化技术，实现从影视元数据到高质量LOGO的全流程自动化生产，极大地缩短了单个影视LOGO的生成时间，降低了影视LOGO的生成成本；同时，通过采用自动化文字审核技术，极大提高了影视LOGO中的文字识别准确率，可满足海量影视内容快速上线的需求，显著提升了平台内容运营效率。
Smart Images
Figure CN122534288A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of artificial intelligence and computer vision technology, and in particular to a method, system, terminal and storage medium for automatically generating film and television logos. Background Technology
[0002] In the daily operation of film and television content platforms, each TV series, movie, or web series needs to be equipped with a unique logo. These logos are not only used in in-platform scenarios such as posters, product detail pages, and recommended placements, but also need to be displayed uniformly on various terminal devices such as mobile devices, tablets, smart TVs, in-vehicle screens, and outdoor advertising screens to strengthen brand recognition and content tone. With the rapid increase in the size of the platform's content library, the workload of logo design has increased exponentially.
[0003] The traditional logo production process relies entirely on professional designers to complete manually, which takes an average of 3-4 hours per logo. If complex Chinese artistic characters or custom graphics are involved, the time will be even longer.
[0004] To alleviate the above problems, a number of AI-assisted design tools have emerged on the market, and their main technical approaches include: 1) Template-based automatic layout: Provides fixed backgrounds and text placeholders; users only need to replace the text. 2) Text Effects Generator: Apply preset filters such as stroke, shadow, and gradient to the input text; 3) Image style transfer: Apply the artistic style of a reference image to text or background.
[0005] However, existing AI tools have the following common shortcomings in actual film and television logo production: (1) Low character recognition rate: Chinese characters have complex structures, many strokes, and many similar characters. Existing tools often treat characters as ordinary image objects, and the generated results often have phenomena such as stroke sticking, missing strokes, excessive ink, and character distortion, resulting in an OCR recognition accuracy rate of generally less than 70%, and even typos, making them completely unusable.
[0006] (2) The style is seriously mismatched with the film and television content: Existing tools lack a deep semantic understanding of film and television metadata (themes, eras, emotions, poster visual elements, etc.).
[0007] (3) Lack of automated review mechanism: After the results are generated, the system relies entirely on manual checks of the text accuracy, style fit, and compliance (whether there are sensitive graphics or inappropriate associations) of each image. There is no built-in quality assessment and filtering capability, which leads to low-quality logos entering the online environment directly, causing user complaints or brand risks.
[0008] (4) The results require a lot of manual secondary adjustments: Even after screening, the candidate logos often have technical flaws such as jagged edges, insufficient contrast between background and text, and abnormal transparency channels. Designers have to go into Photoshop or Illustrator for fine repairs, and each logo still requires more than 30 minutes of extra work on average.
[0009] In summary, the existing technology still has problems such as low recognition rate of generated logo text, mismatch between style and film and television content, and low degree of automation. The technology needs to be improved. Summary of the Invention
[0010] The technical problem to be solved by the present invention is to provide an automated method, system, terminal and storage medium for generating film and television logos, in order to address the shortcomings of the existing technology, such as low text recognition rate of generated logos, style mismatch with film and television content and low degree of automation.
[0011] The technical solution adopted by this invention to solve the technical problem is as follows: In a first aspect, the present invention provides a method for automatically generating film and television logos, including: Obtain the titles and synopsis information of film and television works, and clean up and standardize the special symbols in the titles; Based on the pre-processed title and introduction information, intelligent algorithms are used to generate image generation instructions, text content requirements, font style and background requirements that conform to the film and television style, resulting in prompts for LOGO image generation. Based on the prompt, an image generation model is invoked to create an initial logo image; The initial logo image is subjected to text recognition and comparison to verify the text content contained in the image, and the background of the verified logo image is removed. Upload the optimized logo image to cloud storage and update the task status of the automated movie and TV logo generation.
[0012] In one implementation, obtaining the title and synopsis information of the film / television work, and performing special character cleanup and standardization on the title, includes: The title and synopsis information of the film or television work are obtained through the content interface; Remove special characters from the title and retain the core text content; Key information is extracted from the introduction information to obtain the key information of the film and television works.
[0013] In one implementation, based on the preprocessed title and description information, an intelligent algorithm is used to generate image generation instructions, text content requirements, and font style and background requirements that conform to the film and television style, resulting in prompts for logo image generation, including: Based on the preprocessed title and description information, the image generation instruction is generated using a rule-based decision tree or large language model. The image generation instruction specifies the text content requirements, font style, and background requirements; wherein, the text content requirements are: the generated LOGO image contains the film / TV title text; the font style and background requirements are: a pure black background, and the font style must match the film / TV genre.
[0014] In one implementation, the step of creating an initial logo image based on the prompt word by calling an image generation model includes: The prompt word is input into the image generation system, which processes the film and television logo generation task through an asynchronous processing architecture, periodically polls the task status, and obtains the generated image resources after the task is completed to obtain the initial logo image.
[0015] In one implementation, the step of performing text recognition and comparison on the initial logo image to verify the text content contained in the image, and performing background removal on the verified logo image, includes: The text content in the initial logo image is extracted using a text recognition algorithm, and the recognized text content is compared with the original movie title. When the text recognition accuracy reaches the threshold, the initial LOGO image is deemed to have passed verification. The verified logo image undergoes color processing, and the background is removed from the color channel to obtain the optimized logo image.
[0016] In one implementation, the step of performing color processing on the verified logo image and background masking in the color channels to obtain the optimized logo image includes: Calculate the HSV color space of the verified logo image, and set the saturation channel to 0 to remove all color information; Convert the image after removing color information to grayscale, calculate the histogram curve of the current grayscale image, and find the two highest peaks in the histogram. Color level shifting is performed within the two peaks, and gamma correction mapping is applied to obtain the adjusted grayscale image. The adjusted grayscale image is used as the Alpha channel, and the original colors of the image are filled into the RGB channels to obtain the optimized LOGO image.
[0017] In one implementation, uploading the optimized logo image to cloud storage and updating the task status of the automated movie logo generation includes: The distribution density of semi-transparent pixels in the transparency channel of the optimized logo image is analyzed, and the edge area of a preset width around the image is scanned to obtain the detection results of color transparency rationality and edge integrity. Based on the detection results, the initial logo image and the optimized logo image are uploaded to cloud storage respectively, key parameters in the generation process are recorded, and the task status of the automated generation of the film and television logo is updated.
[0018] Secondly, the present invention provides an automated system for generating film and television logos, comprising: The metadata acquisition and preprocessing module is used to acquire the title and synopsis information of film and television works, and to clean up and standardize special characters in the title. The intelligent prompt word generation module is used to generate prompt words for LOGO image generation based on the pre-processed title and introduction information, using intelligent algorithms to generate image generation instructions, text content requirements, font style and background requirements that conform to the film and television style. The intelligent logo image generation module is used to create an initial logo image by calling an image generation model based on the prompt words; An automated text verification module is used to perform text recognition and comparison on the initial LOGO image to verify the text content contained in the image; The image post-processing optimization module is used to remove the background from verified logo images; The results storage and status update module is used to upload the optimized logo image to cloud storage and update the task status of automated movie logo generation.
[0019] Thirdly, the present invention provides a terminal, comprising: a processor and a memory, wherein the memory stores a film and television logo automatic generation program, and the film and television logo automatic generation program, when executed by the processor, is used to implement the operation of the film and television logo automatic generation method as described in the first aspect.
[0020] Fourthly, the present invention also provides a computer-readable storage medium storing a film and television logo automatic generation program, which, when executed by a processor, is used to implement the film and television logo automatic generation method as described in the first aspect.
[0021] The present invention, by employing the above technical solution, has the following effects: By integrating language understanding and image generation capabilities, combining automated review and post-optimization technologies, the present invention realizes the full-process automated production from film and television metadata to high-quality LOGOs, greatly shortening the generation time of individual film and television LOGOs and reducing the generation cost of film and television LOGOs. At the same time, by adopting automated text review technology, the accuracy of text recognition in film and television LOGOs is greatly improved, meeting the needs of rapid online launch of a large amount of film and television content and significantly enhancing the platform content operation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on the structures shown in these drawings.
[0023] Figure 1 It is a flowchart of the method for automatically generating film and television LOGOs in the present invention.
[0024] Figure 2 It is the overall flowchart of the method for automatically generating film and television LOGOs in the present invention.
[0025] Figure 3 It is the functional schematic diagram of the terminal in one implementation manner of the present invention.
[0026] The realization of the object of the present invention, functional features and advantages will be further described with reference to the embodiments and the drawings. SPECIFIC EMBODIMENTS
[0027] To make the object, technical solution and advantages of the present invention clearer and more definite, the following will further describe the present invention in detail with reference to the accompanying drawings and by way of examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0028] EXEMPLARY METHOD The following are the common defects of existing AI tools in the actual production of film and television LOGOs: (1) Low text recognition rate: Chinese characters have complex structures, many strokes and many similar characters. Existing tools often regard text as ordinary image objects, and phenomena such as stroke adhesion, missing strokes, excessive ink, and distorted glyphs are common in the generated results, resulting in an OCR recognition accuracy rate generally lower than 70%, and even typos (such as "Wolf Warrior" becoming "Wolf Savage"), making it completely unusable directly.
[0029] (2) Severe mismatch between style and film / television content: Existing tools lack a deep semantic understanding of film and television metadata (theme, era, mood, poster visual elements, etc.). For example, automatically generating a trendy variety show-style colorful font for a period suspense drama, or adding a serious bold font to an animated film. Designers still need to spend a lot of time adjusting parameters such as font family, weight, and color, and the actual time saved is limited.
[0030] (3) Lack of automated review mechanism: After the results are generated, the system relies entirely on manual checks of the text accuracy, style fit, and compliance (whether there are sensitive graphics or inappropriate associations) of each image. There is no built-in quality assessment and filtering capability, which leads to low-quality logos entering the online environment directly, causing user complaints or brand risks.
[0031] (4) The results require a lot of manual secondary adjustments: Even after screening, the candidate logos often have technical flaws such as jagged edges, insufficient contrast between background and text, and abnormal transparency channels. Designers have to go into Photoshop or Illustrator for fine repairs, and each logo still requires more than 30 minutes of extra work on average.
[0032] In summary, the existing technology still has problems such as low recognition rate of generated logo text, mismatch between style and film and television content, and low degree of automation. The technology needs to be improved.
[0033] To address the above technical issues, this invention provides an automated method for generating film and television logos. The method includes: acquiring the title and synopsis information of the film or television work; cleaning and standardizing the title for special symbols; generating image generation instructions, text content requirements, font style, and background requirements that conform to the film and television style using an intelligent algorithm, resulting in prompts for logo image generation; creating an initial logo image by calling an image generation model based on the prompts; performing text recognition and comparison on the initial logo image to verify the text content contained in the image, and performing background cutout on the verified logo image; uploading the optimized logo image to cloud storage and updating the task status of the automated film and television logo generation. This invention achieves fully automated production from film and television metadata to high-quality logo images.
[0034] like Figure 1 As shown, this embodiment of the invention provides a method for automatically generating film and television logos, including the following steps: Step S100: Obtain the title and synopsis information of the film and television works, and clean up and standardize the special symbols in the title.
[0035] In this embodiment, an automated method for generating film and television logos is proposed. By integrating language understanding and image generation capabilities, and combining automated review and post-production optimization technologies, the method achieves fully automated production from film and television metadata to high-quality logos.
[0036] like Figure 2 As shown, the automated method for generating film and television logos mainly includes the following key stages: 1) Metadata Acquisition and Preprocessing: Obtain the title and synopsis information of film and television works from the content management system, and clean up and standardize the special symbols in the title.
[0037] 2) Intelligent prompt word generation: Based on film and television metadata, intelligent algorithms generate image generation instructions that conform to the film and television style, and clearly specify the text content, font style and background requirements.
[0038] 3) Intelligent generation of logo image: Based on the generated prompts, the image generation model is called to create an initial logo image.
[0039] 4) Automated text verification: The generated logo is subjected to text recognition and comparison to verify whether it contains the correct text content.
[0040] 5) Image post-processing optimization: Background removal for approved logos.
[0041] 6) Result storage and status update: Upload the optimized logo to cloud storage and update the task status.
[0042] Specifically, in one implementation of this embodiment, step S100 includes the following steps: Step S101: Obtain the title and synopsis information of the film or television work through the content interface; Step S102: Clean up the special characters in the title and retain the core text content; Step S103: Extract key information from the introduction information to obtain the key information of the film and television work.
[0043] In this embodiment, during the metadata acquisition and preprocessing stage, the title and synopsis of the specified film and television work are first obtained through a content interface (e.g., TMDB, Douban API, or internal media asset library). Then, special characters in the title are cleaned up, retaining the core text content. Specifically, redundant spaces, line breaks, emojis, and special symbols (e.g., 【】, #, etc.) are removed, retaining only Chinese and English characters, numbers, and common punctuation marks to ensure the title is concise and standardized. Finally, key information is extracted from the synopsis. For example, regular expression matching or lightweight NLP methods (such as TF-IDF and TextRank methods) are used to extract core elements (including main characters, story background, genre keywords, emotional tone, etc.). The cleaned title and the extracted key information are then stored in a structured manner (e.g., in JSON format) to provide an accurate and low-noise input source for subsequent prompt word generation.
[0044] like Figure 1 As shown, this embodiment of the invention provides a method for automatically generating film and television logos, including the following steps: Step S200: Based on the preprocessed title and introduction information, an intelligent algorithm is used to generate image generation instructions, text content requirements, font style and background requirements that conform to the film and television style, and obtain prompts for LOGO image generation.
[0045] In this embodiment, during the intelligent prompt word generation stage, an image generation instruction is generated based on the preprocessed film and television metadata using an intelligent algorithm. The instruction explicitly requires the generation of a logo containing the film and television title text and specifies a pure black background for subsequent image cutout processing. The instruction also describes that the font style must match the film and television genre (such as suspense, romance, science fiction, etc.).
[0046] Specifically, in one implementation of this embodiment, step S200 includes the following steps: Step S201: Based on the preprocessed title and description information, generate the image generation instruction using a rule-based decision tree or large language model; Step S202: In the image generation instruction, the text content requirements and font style and background requirements are determined; wherein, the text content requirements are: the generated LOGO image contains the film / TV title text; the font style and background requirements are: pure black background, and the font style must match the film / TV type.
[0047] In this embodiment, based on the video metadata (cleaned title + key information) output in the preprocessing stage, instructions for image generation models (e.g., mixed text-to-image large model) are dynamically generated using preset prompt word templates or a lightweight rule engine. These instructions specify the following three generation elements: 1) It is explicitly required that a logo containing the film / TV title be generated. The text must be clear and legible and serve as one of the core visual elements. 2) Specify the background as pure black (#000000) to facilitate precise image cutout processing later using chroma keying or color range selection; 3) Select the corresponding font and visual effect description from the style mapping table according to the film and television genre (suspense, romance, science fiction, etc.). For example, suspense: rough, slanted, shadow, fragmented; romance: rounded, handwritten, soft halo; science fiction: geometric lines, metallic texture, neon light effect.
[0048] In the intelligent prompt generation stage of this embodiment, the final output prompt is a natural language paragraph, which includes constraints such as screen composition, text position, font style, and pure black background, and is accompanied by necessary negative prompts (e.g., "no background with mixed colors, no complex textures") to improve generation stability.
[0049] like Figure 1 As shown, this embodiment of the invention provides a method for automatically generating film and television logos, including the following steps: Step S300: Based on the prompt word, call the image generation model to create an initial LOGO image.
[0050] In this embodiment, during the logo image generation stage, the generated prompt words are submitted to the image generation system. Then, the system asynchronously processes the generation task, periodically polls the task status, and obtains the generated image resources after the task is completed.
[0051] Specifically, in one implementation of this embodiment, step S300 includes the following steps: Step S301: Input the prompt word into the image generation system. The image generation system processes the film and television logo generation task through an asynchronous processing architecture, periodically polls the task status, and obtains the generated image resources after the task is completed to obtain the initial logo image.
[0052] In this embodiment, the prompt words generated in the previous step are input into the image generation model; during the LOGO image generation stage, the system adopts an asynchronous processing method, and the client polls the task status interface at fixed intervals (e.g., 2 seconds) until the status returns "success" or "failure"; to avoid infinite polling, a timeout period (e.g., 60 seconds) and a maximum number of retries are set, and the task is marked as failed and logged after the timeout.
[0053] After the logo image generation task is completed, the generated logo image (in PNG format, with reserved transparency channel) is downloaded through the result interface and stored in a temporary directory or object storage, while the image generation parameters are recorded.
[0054] like Figure 1 As shown, this embodiment of the invention provides a method for automatically generating film and television logos, including the following steps: Step S400: Perform text recognition and comparison on the initial LOGO image to verify the text content contained in the image, and perform background removal on the verified LOGO image.
[0055] In this embodiment, during the automated review stage, the system downloads the generated logo image, extracts the text content from the image using a text recognition algorithm, and compares the recognition result with the original movie title. Only when the text recognition accuracy reaches a threshold is the review deemed passed. For the approved logo image, it is optimized through an image refinement process.
[0056] Specifically, in one implementation of this embodiment, step S400 includes the following steps: Step S401: Extract the text content from the initial LOGO image using a text recognition algorithm, and compare the recognized text content with the original movie title; Step S402: When the text recognition accuracy reaches the threshold, the initial LOGO image is determined to have passed verification.
[0057] In this embodiment, the generated LOGO image is downloaded from the local cache or object storage and decoded into memory format. Then, a text recognition algorithm (e.g., PaddleOCR or cloud service OCR) is called to extract all text regions and their contents from the image. Finally, the recognized text is matched with the original movie title. During the matching process, a text recognition accuracy threshold can be set (e.g., threshold > 90%). Only when the matching result is higher than the threshold is the review considered passed; otherwise, manual review or regeneration process is triggered.
[0058] For logos that fail the review, the reason for the failure is recorded and a retry is allowed; for logos that pass the review, the following logo optimization processing method is performed.
[0059] Step S403: Perform color processing on the verified LOGO image and perform background removal in the color channel to obtain the optimized LOGO image.
[0060] In one implementation of this embodiment, step S403 includes the following steps: Step S403a: Calculate the HSV color space of the verified LOGO image, and set the saturation channel to 0 to remove all color information; Step S403b: Convert the image after removing color information into a grayscale image, calculate the histogram curve of the current grayscale image, and find the two highest peaks in the histogram. Step S403c: Perform color level shifting processing within the two peaks and perform gamma correction mapping to obtain the adjusted grayscale image; Step S403d: The adjusted grayscale image is used as the Alpha channel, and the original colors of the image are filled into the RGB channels to obtain the optimized LOGO image.
[0061] In this embodiment, the image refinement process in the post-image optimization stage is as follows: First, the image's HSV color space is calculated, and the S (saturation) channel is set to 0, thus removing all color information. Next, the color-removed image is converted to grayscale, and its histogram is calculated. The two highest peaks in the histogram are identified; these peaks primarily represent the color distribution of the background and the subject. Finally, within these two peaks, the color levels are shifted to 20% of the two peaks (the total color level range is 0-255; for example, if the two peaks are 100 and 200, shifting inward by 20% moves the color level range to 120-180), and gamma correction is applied to map it to the 0-255 range to enhance the logo's edge smoothness.
[0062] The method for generating a logo image by compositing channels is: Alpha channel + RGB channel = finished logo image.
[0063] 1) Alpha channel: Use an adjusted grayscale image directly as the alpha channel to achieve a gradient effect from completely transparent to opaque.
[0064] 2) RGB channel: Fill the original color of the image to get the optimized logo image.
[0065] In this embodiment, by combining the generation of a specified solid color background with intelligent image cutout, the precise separation of the film and television logo from the background is achieved; furthermore, the optimal color level parameters are automatically calculated and adjusted based on the image content, significantly improving the visibility of the logo under different backgrounds.
[0066] like Figure 1 As shown, this embodiment of the invention provides a method for automatically generating film and television logos, including the following steps: Step S500: Upload the optimized logo image to cloud storage and update the task status of automated movie logo generation.
[0067] In this embodiment, the optimized logo image is finally tested using color transparency and edge integrity detection methods to ensure that the generated logo image meets the quality requirements.
[0068] Specifically, in one implementation of this embodiment, step S500 includes the following steps: Step S501: Analyze the distribution density of semi-transparent pixels in the transparency channel of the optimized LOGO image, and scan the edge area of the image with a preset width to obtain the detection results of color transparency rationality and edge integrity. Step S502: Based on the detection results, upload the initial LOGO image and the optimized LOGO image to cloud storage respectively, record the key parameters in the generation process, and update the task status of the automated generation of the film and television LOGO.
[0069] In this embodiment, the detection principle of the color transparency reasonableness detection method is as follows: analyze the distribution density of semi-transparent pixels in the transparency channel. The color transparency reasonableness judgment logic is as follows: if the proportion of semi-transparent area exceeds a preset safety threshold, it is judged as abnormal color transparency (such as excessive glass texture), triggering a reprocessing process.
[0070] The edge integrity detection method works by scanning the edge area of a preset width around the image. The edge integrity judgment logic is as follows: if non-transparent pixels exist in the edge area, it is determined that the logo body is not properly aligned with the edge or that the cropping is abnormal, and is therefore considered unacceptable.
[0071] For logo images that pass the color transparency and edge integrity checks, the results are stored. In this stage, the initially generated logo image and the optimized logo image are uploaded to cloud storage, and key parameters during the generation process (such as prompts, processing status, etc.) are recorded. Finally, the task status is updated to complete or failed, and detailed processing information is provided.
[0072] This embodiment achieves the following technical effects through the above technical solution: This invention integrates language understanding and image generation capabilities, combined with automated review and post-production optimization technologies, to achieve fully automated production from film and television metadata to high-quality logos. This significantly shortens the generation time and reduces the cost of producing individual film and television logos. At the same time, by adopting automated text review technology, the accuracy of text recognition in film and television logos is greatly improved, which can meet the needs of rapid online release of massive amounts of film and television content and significantly improve the platform's content operation efficiency.
[0073] Exemplary device Based on the above embodiments, the present invention also provides an automated film and television logo generation system, comprising: The metadata acquisition and preprocessing module is used to acquire the title and synopsis information of film and television works, and to clean up and standardize special characters in the title. The intelligent prompt word generation module is used to generate prompt words for LOGO image generation based on the pre-processed title and introduction information, using intelligent algorithms to generate image generation instructions, text content requirements, font style and background requirements that conform to the film and television style. The intelligent logo image generation module is used to create an initial logo image by calling an image generation model based on the prompt words; An automated text verification module is used to perform text recognition and comparison on the initial LOGO image to verify the text content contained in the image; The image post-processing optimization module is used to remove the background from verified logo images; The results storage and status update module is used to upload the optimized logo image to cloud storage and update the task status of automated movie logo generation.
[0074] Based on the above embodiments, the present invention also provides a terminal, the principle block diagram of which can be as follows: Figure 3 As shown.
[0075] The terminal includes: a processor, a memory, an interface, a display screen, and a communication module connected via a system bus; wherein, the processor of the terminal provides computing and control capabilities; the memory of the terminal includes a computer-readable storage medium and internal memory; the computer-readable storage medium stores an operating system and computer programs; the internal memory provides an environment for the operation of the operating system and computer programs in the computer-readable storage medium; the interface is used to connect to external devices; the display screen is used to display relevant information; and the communication module is used to communicate with a cloud server or other devices.
[0076] When executed by a processor, this computer program is used to implement the automated generation method for film and television logos.
[0077] It will be understood by those skilled in the art that Figure 3 The schematic diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal to which the present invention is applied. A specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0078] In one embodiment, a terminal is provided, comprising: a processor and a memory, the memory storing a film and television logo automatic generation program, which, when executed by the processor, is used to implement the above-described film and television logo automatic generation method.
[0079] In one embodiment, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a film and television logo automatic generation program, which, when executed by a processor, is used to implement the above-described film and television logo automatic generation method.
[0080] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, database, or other media used in the embodiments provided by this invention can include both non-volatile and volatile memory.
[0081] In summary, this invention provides a method, system, terminal, and storage medium for automated generation of film and television logos, including: acquiring the title and synopsis information of the film and television work, and cleaning and standardizing the special symbols in the title; based on the preprocessed title and synopsis information, generating image generation instructions, text content requirements, font style, and background requirements that conform to the film and television style through an intelligent algorithm to obtain prompts for logo image generation; based on the prompts, calling the image generation model to create an initial logo image; performing text recognition and comparison on the initial logo image to verify the text content contained in the image, and performing background removal on the verified logo image; uploading the optimized logo image to cloud storage and updating the task status of automated film and television logo generation; this invention realizes fully automated production from film and television metadata to high-quality logo images.
[0082] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.
Claims
1. A method for automatically generating film and television logos, characterized in that, include: Obtain the titles and synopsis information of film and television works, and clean up and standardize the special symbols in the titles; Based on the pre-processed title and introduction information, intelligent algorithms are used to generate image generation instructions, text content requirements, font style and background requirements that conform to the film and television style, resulting in prompts for LOGO image generation. Based on the prompt, an image generation model is invoked to create an initial logo image; The initial logo image is subjected to text recognition and comparison to verify the text content contained in the image, and the background of the verified logo image is removed. Upload the optimized logo image to cloud storage and update the task status of the automated movie and TV logo generation.
2. The automated method for generating film and television logos according to claim 1, characterized in that, The process of obtaining the title and synopsis information of film and television works, and cleaning up and standardizing special characters in the titles, includes: The title and synopsis information of the film or television work are obtained through the content interface; Remove special characters from the title and retain the core text content; Key information is extracted from the introduction information to obtain the key information of the film and television works.
3. The automated method for generating film and television logos according to claim 1, characterized in that, Based on the preprocessed title and introduction information, an intelligent algorithm generates image generation instructions, text content requirements, and font and background requirements that conform to the film and television style, resulting in prompts for logo image generation, including: Based on the preprocessed title and description information, the image generation instruction is generated using a rule-based decision tree or large language model. The image generation instruction specifies the text content requirements, font style, and background requirements; wherein, the text content requirements are: the generated LOGO image contains the film / TV title text; the font style and background requirements are: a pure black background, and the font style must match the film / TV genre.
4. The automated method for generating film and television logos according to claim 1, characterized in that, The step of creating an initial logo image by calling an image generation model based on the prompt word includes: The prompt word is input into the image generation system, which processes the film and television logo generation task through an asynchronous processing architecture, periodically polls the task status, and obtains the generated image resources after the task is completed to obtain the initial logo image.
5. The automated method for generating film and television logos according to claim 1, characterized in that, The step of performing text recognition and comparison on the initial logo image to verify the text content contained in the image, and performing background removal on the verified logo image, includes: The text content in the initial logo image is extracted using a text recognition algorithm, and the recognized text content is compared with the original movie title. When the text recognition accuracy reaches the threshold, the initial LOGO image is deemed to have passed verification. The verified logo image undergoes color processing, and the background is removed from the color channel to obtain the optimized logo image.
6. The automated method for generating film and television logos according to claim 5, characterized in that, The process of color processing the verified logo image and performing background removal in the color channels to obtain the optimized logo image includes: Calculate the HSV color space of the verified logo image, and set the saturation channel to 0 to remove all color information; Convert the image after removing color information to grayscale, calculate the histogram curve of the current grayscale image, and find the two highest peaks in the histogram. Color level shifting is performed within the two peaks, and gamma correction mapping is applied to obtain the adjusted grayscale image. The adjusted grayscale image is used as the Alpha channel, and the original colors of the image are filled into the RGB channels to obtain the optimized LOGO image.
7. The automated method for generating film and television logos according to claim 1, characterized in that, The step of uploading the optimized logo image to cloud storage and updating the task status of the automated movie logo generation includes: The distribution density of semi-transparent pixels in the transparency channel of the optimized logo image is analyzed, and the edge area of a preset width around the image is scanned to obtain the detection results of color transparency rationality and edge integrity. Based on the detection results, the initial logo image and the optimized logo image are uploaded to cloud storage respectively, key parameters in the generation process are recorded, and the task status of the automated generation of the film and television logo is updated.
8. An automated logo generation system for film and television production, characterized in that, include: The metadata acquisition and preprocessing module is used to acquire the title and synopsis information of film and television works, and to clean up and standardize special characters in the title. The intelligent prompt word generation module is used to generate prompt words for LOGO image generation based on the pre-processed title and introduction information, using intelligent algorithms to generate image generation instructions, text content requirements, font style and background requirements that conform to the film and television style. The intelligent logo image generation module is used to create an initial logo image by calling an image generation model based on the prompt words; An automated text verification module is used to perform text recognition and comparison on the initial LOGO image to verify the text content contained in the image; The image post-processing optimization module is used to remove the background from verified logo images; The results storage and status update module is used to upload the optimized logo image to cloud storage and update the task status of automated movie logo generation.
9. A terminal, characterized in that, include: The processor and memory, wherein the memory stores a film and television logo automatic generation program, which, when executed by the processor, is used to implement the operation of the film and television logo automatic generation method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a film and television logo automatic generation program, which, when executed by a processor, is used to implement the operation of the film and television logo automatic generation method as described in any one of claims 1-7.