Control method and control system for car lamp projection

Abstract

The invention discloses a control method and a control system for car lamp projection. The control method comprises the following steps: acquiring an initial text to be projected according to voice input; whether the initial text comprises preset standard information or not is judged, and the preset standard information comprises projection keywords and law and regulation pattern names; in response to the situation that the initial text does not comprise the preset standard information, generating a first projection pattern based on the initial text through a text generation graph large model; and at least post-processing the first projection pattern, so that the first projection pattern is projected and displayed in a distortionless manner according to vehicle lamp pixels. By executing the steps, the process of car lamp projection can be simpler, more convenient and quicker, meanwhile, personalized car lamp projection patterns can be created according to the imagination of a user, the limitation of the prefabricated content of car lamp projection is broken through, the user experience is enhanced, and the interactivity, diversity and playability of the car lamp projection content are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile headlight display, and in particular to a method for controlling headlight projection, a control system for headlight projection, and a computer-readable storage medium. Background Art

[0002] Currently, the images projected by headlights are often pre-built into the vehicle's computer system by a third party. For example, car manufacturers can pre-store existing images in the vehicle's computer system database before the car leaves the factory. Users can also proactively store their own frequently used images in the database based on their daily use of headlight projection.

[0003] In addition, in order to simplify the operational process of car light projection, the existing technology also provides a projection method that can control the projection content of the car light through user voice input. Based on the user's description of the projection object, the image that meets the corresponding description is retrieved from the database, and then called and output for projection display. Furthermore, if there is no image that meets the user's description in the current database, keywords can be extracted from the user's description, and then feedback data matching the keywords corresponding to the description text is obtained from the material library. The feedback data and its corresponding keywords are fed back to the database to prepare a new image for updating the database.

[0004] However, regardless of the aforementioned methods, the image to be projected must be pre-stored or pre-prepared in a database. The driver or passenger then clicks on the pre-prepared image on the central control screen, which is then projected by the headlights. In other words, existing headlight projection methods can only project and display images pre-prepared in the vehicle's computer system. This results in limited functionality, poor playability, and significantly limits the user's creativity and personalized needs for the desired projection pattern.

[0005] In order to solve the above-mentioned problems existing in the prior art, this field urgently needs a control technology for car light projection, which can not only make the car light projection process simpler and faster, but also create personalized car light projection patterns according to the user's imagination, breaking through the limitations of pre-made content of car light projection, enhancing user experience, and improving the interactivity, diversity and playability of car light projection content. Summary of the Invention

[0006] The following is a brief summary of one or more aspects to provide a basic understanding of these aspects. This summary is not an exhaustive overview of all conceivable aspects and is neither intended to identify key or critical elements of all aspects nor to define the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that will be provided later.

[0007] In order to overcome the above-mentioned defects of the prior art, the present invention provides a control method for car light projection, a control system for car light projection, and a computer-readable storage medium, which not only make the process of car light projection simpler and faster, but also can create personalized car light projection patterns according to the user's imagination, breaking through the limitations of pre-made content of car light projection, enhancing user experience, and improving the interactivity, diversity and playability of car light projection content.

[0008] Specifically, the above-mentioned control method for headlight projection provided according to the first aspect of the present invention includes the following steps: obtaining an initial text to be projected based on voice input; judging whether the initial text includes preset standard information, wherein the preset standard information includes projection keywords and regulatory pattern names; in response to the inclusion of projection keywords in the initial text, generating a first projection pattern of a guidance object based on the projection keywords through a large model of a cultural map; in response to the fact that the initial text does not include the preset standard information, generating a second projection pattern based on the initial text through the large model of a cultural map; and post-processing at least the first projection pattern and / or the second projection pattern so that the first projection pattern and / or the second projection pattern are projected and displayed without distortion according to the headlight pixels.

[0009] In addition, the control system for the above-mentioned headlight projection provided according to the second aspect of the present invention includes: a vehicle system, which is used to obtain an initial text to be projected based on voice input; determine whether the initial text includes preset standard information, wherein the preset standard information includes projection keywords and regulatory pattern names; a computing platform, which is communicatively connected to the vehicle system, includes a large model of a cultural image, which is used to include projection keywords in the initial text, and generate a first projection pattern of a guidance object based on the projection keywords through the large model of the cultural image; when the initial text does not include the preset standard information, generate a second projection pattern based on the initial text through the large model of the cultural image; and return it to the vehicle system, so that the vehicle system at least post-processes the first projection pattern and / or the second projection image; a headlight electronic control unit, which is communicatively connected to the vehicle system, at least for obtaining the first projection pattern after post-processing; and a headlight module, which is communicatively connected to the headlight electronic control unit, for projecting and displaying the first projection pattern after post-processing according to the headlight pixels without distortion.

[0010] Furthermore, according to a third aspect of the present invention, there is provided a computer-readable storage medium having computer instructions stored thereon, wherein when the computer instructions are executed by a processor, the vehicle light projection control method provided in the first aspect of the present invention is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above features and advantages of the present invention will be better understood after reading the detailed description of the embodiments of the present disclosure in conjunction with the following drawings. In the drawings, the components are not necessarily drawn to scale, and components with similar related properties or characteristics may have the same or similar reference numerals.

[0012] Figure 1 A schematic diagram showing the layout of a control system for vehicle headlight projection according to some embodiments of the present invention is shown;

[0013] Figure 2 A flow chart of a method for controlling vehicle light projection according to some embodiments of the present invention is shown;

[0014] Figure 3 A module block diagram of a vehicle light projection control system according to some embodiments of the present invention is shown;

[0015] Figure 4 A schematic diagram showing a process of adjusting the size of a projection pattern according to some embodiments of the present invention is shown; and

[0016] Figure 5 A schematic diagram of a process of performing trapezoidal correction on an original pre-projection pattern according to some embodiments of the present invention is shown.

[0017] Reference numerals:

[0018] 100 Control system for headlight projection;

[0019] 110 Car computer system;

[0020] 120 computing platform;

[0021] 121 cloud server platform;

[0022] 122 local computing platform;

[0023] 130 headlight electronic control unit;

[0024] 140 headlight modules; and

[0025] Steps S210 to S250. DETAILED DESCRIPTION

[0026] The following specific embodiments illustrate the embodiments of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Although the description of the present invention will be introduced in conjunction with the preferred embodiment, this does not mean that the features of this invention are limited to this embodiment. On the contrary, the purpose of introducing the invention in conjunction with the embodiment is to cover other options or modifications that may be extended based on the claims of the present invention. In order to provide a deep understanding of the present invention, the following description will include many specific details. The present invention can also be implemented without using these details. In addition, in order to avoid confusion or blurring the focus of the present invention, some specific details will be omitted in the description.

[0027] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0028] Furthermore, the terms "upper," "lower," "left," "right," "top," "bottom," "horizontal," and "vertical" used in the following description should be understood to refer to the orientations depicted in that section and the accompanying drawings. These relative terms are used solely for convenience of description and do not necessarily imply that the devices described herein must be manufactured or operated in a specific orientation. Therefore, they should not be construed as limiting the present invention.

[0029] It will be understood that although the terms "first," "second," "third," etc. may be used herein to describe various components, regions, layers, and / or portions, these components, regions, layers, and / or portions should not be limited by these terms, and these terms are merely used to distinguish different components, regions, layers, and / or portions. Thus, a first component, region, layer, and / or portion discussed below may be referred to as a second component, region, layer, and / or portion without departing from some embodiments of the present invention.

[0030] As mentioned above, existing technologies require that the patterns to be projected be pre-stored or pre-prepared in a database. The driver or passenger then clicks on the pre-prepared images on the central control screen, which then project the images through the headlights. In other words, existing headlight projection methods can only project patterns pre-prepared in the vehicle's computer system, resulting in limited functionality and limited playability. Furthermore, these methods significantly limit the user's creativity and personalized needs for the desired projection patterns.

[0031] In order to solve the above-mentioned problems existing in the prior art, the present invention provides a control method for car light projection, a control system for car light projection, and a computer-readable storage medium, which not only make the process of car light projection simpler and faster, but also can create personalized car light projection patterns according to the user's imagination, breaking through the limitations of pre-made content of car light projection, enhancing user experience, and improving the interactivity, diversity and playability of car light projection content.

[0032] In some non-limiting embodiments, the vehicle light projection control method provided by the first aspect of the present invention may be implemented via the vehicle light projection control system provided by the second aspect of the present invention.

[0033] Specifically, in some non-limiting embodiments, the computer-readable storage medium provided in the third aspect of the present invention stores computer instructions, which, when executed by a processor, can be used to implement the vehicle light projection control method provided in the first aspect of the present invention.

[0034] The following describes the specific implementation of the aforementioned vehicle projection control method, using several examples of vehicle projection control systems. Those skilled in the art will appreciate that these examples of vehicle projection control systems are merely non-limiting implementations of the present invention, intended to clearly illustrate the main concepts of the present invention and provide specific solutions for easy implementation. They do not restrict the implementation of the various steps in these vehicle projection control methods. Similarly, the vehicle projection control method is also intended to limit the entire operating mode or functionality of the vehicle projection control system.

[0035] Please see Figure 1 , Figure 1 A schematic diagram of the layout of a control system for vehicle light projection provided in accordance with some embodiments of the present invention is shown.

[0036] like Figure 1 As shown, in some embodiments of the present invention, a vehicle with projection functionality may include a headlight projection control system 100. Specifically, headlight projection control system 100 may include a vehicle-mounted system 110 located within the vehicle, a headlight electronic control unit 130 (referred to as the headlight ECU), and a headlight module 140. Headlight module 140 may include a headlight array composed of multiple rows of LED lights. Headlight module 140 may communicate with headlight ECU 130 via UART-CAN and LVDS.

[0037] Furthermore, the control system 100 for vehicle light projection may further include a computing platform 120 for providing a plurality of different types of large models to process data information transmitted from the vehicle system 110 .

[0038] like Figure 1 As shown, the computing platform 120 can optionally be a local computing platform 122. The local computing platform 122 can also be deployed in the vehicle and can be connected to the vehicle system 110 and the headlight electronic control unit 130 via Ethernet. Because the local computing platform 122 is a separate computing platform installed in each vehicle and can specifically serve a single vehicle system, the model operation speed is faster, but the configuration is also higher.

[0039] In other optional embodiments, the computing platform 120 may also utilize a cloud server platform 121. The cloud server platform 121 can communicate with the vehicle-mounted system 110 via a 4G / 5G network. Furthermore, the cloud server platform 121 can connect to multiple vehicle-mounted systems 110 in multiple vehicles, allowing them to share large model resources. This can reduce the configuration cost of the computing platform 120, but the model operation speed will be reduced.

[0040] Next, please combine Figure 2 and Figure 3 It is commonly understood that the specific process of the vehicle light projection control method provided by the present invention. Figure 2 FIG2 shows a module block diagram of a control system for vehicle light projection according to some embodiments of the present invention. Figure 3 A flow chart of a method for controlling vehicle light projection according to some embodiments of the present invention is shown.

[0041] like Figure 2 As shown, in some embodiments of the present invention, the control method of vehicle light projection may include the following steps: First, step 210 may be performed: obtaining an initial text to be projected according to a voice input.

[0042] Specifically, combined Figure 3 As shown, in some embodiments, the vehicle-mounted system 110 may include a voice input module and a voice recognition module. When a driver or passenger describes a desired content projected by the headlights in the vehicle cabin, the voice input module in the vehicle-mounted system 110, such as a microphone or other audio receiving device, may capture the user's voice input information. The voice input information may include at least the content projected by the headlights. For example, the voice input module may capture multiple voice input messages from the user in the vehicle, only one of which is relevant to the content projected by the headlights.

[0043] like Figure 3As shown, after the vehicle system 110 receives the voice input information from the vehicle occupant, the voice input information can be parsed into corresponding coded data through the voice recognition module and uploaded to the computing platform 120. The voice recognition algorithm of the cloud server platform 121 or the local computing platform 122 is called to perform voice recognition, generate a voice recognition result, and return the result to the voice recognition module of the vehicle system 110.

[0044] Specifically, the principles of the speech recognition algorithm in the cloud server platform 121 or the local computing platform 122 can primarily include the following key steps. First, the speech input signal can be preprocessed. Because speech input signals are often affected by various noise and interference, preprocessing is required. Specifically, this can include sampling and quantizing the raw speech signal, converting it into a digital signal for computer processing. Furthermore, filtering and denoising operations can be performed to improve the quality of the speech signal, reduce the impact of background noise, and enhance the characteristics of the speech.

[0045] After this, feature extraction can be performed on the preprocessed speech input signal. Specifically, parameters representing speech characteristics can be extracted from the preprocessed speech input signal. Commonly used feature parameters include Mel-Frequency Cepstral Coefficients (MFCCs) and Linear Predictive Coding (LPC) coefficients. Taking MFCCs as an example, they are based on the human auditory characteristics and logarithmically compress and filter the energy of the speech signal at different frequencies to obtain a set of coefficients that reflect the spectral characteristics of the speech. These feature parameters can effectively describe the acoustic characteristics of speech and provide a foundation for the subsequent recognition process.

[0046] After that, acoustic model training can be performed. The acoustic model is one of the core parts of the speech recognition system. It can be used to map the extracted speech features to the corresponding phonemes or words. Common acoustic models can include the hidden Markov model (HMM) and its variants, such as the deep neural network hidden Markov model (DNN-HMM). During the training process, a large amount of labeled speech data is required, and the parameters of the model are adjusted through the optimization algorithm so that the model can predict the acoustic units corresponding to the speech features as accurately as possible. Taking HMM as an example, it regards the speech signal as a random process consisting of a series of hidden states and observations. By learning the transition probability between states and the probability distribution of the observations under each state, it establishes the mapping relationship between speech features and acoustic units.

[0047] Subsequently, a language model can be constructed to constrain and optimize the output of the acoustic model based on contextual information, thereby improving the accuracy and naturalness of recognition. The language model is mainly used to describe the probability of occurrence and grammatical structure of words or sentences in a language. It can help the speech recognition algorithm select the result that best conforms to the language rules and semantics among multiple possible recognition results. Commonly used language models may include N-gram models and neural network language models (such as the recurrent neural network language model RNN-LM, the long short-term memory network language model LSTM-LM, etc.). Among them, the N-gram model is based on a statistical method, which estimates the probability of a sentence by calculating the frequency of N adjacent words appearing at the same time, while the neural network language model can more effectively process long sequences of language information and learn more complex language patterns and semantic relationships.

[0048] During the recognition phase, the language model can be combined with the acoustic model to constrain the acoustic model's output. For example, if the acoustic model recognizes an unlikely word sequence, the language model can adjust it to make it more consistent with linguistic patterns. The extracted speech features are fed into the acoustic and language models, and a decoding search algorithm is used to find the most likely recognition result. Common decoding algorithms include the Viterbi algorithm. The Viterbi algorithm is a dynamic programming algorithm that searches for the optimal path in the joint probability space of the acoustic and language models—that is, the word or sentence sequence with the highest probability. This algorithm calculates the probabilities of all possible states at each time step and selects the path with the highest probability for expansion, ultimately achieving the optimal recognition result for the entire speech sequence.

[0049] With the development of deep learning technology, end-to-end speech recognition algorithms have been widely researched and applied. These algorithms directly map speech signals to text without explicitly building acoustic and language models. Instead, they automatically learn speech feature representations and semantic information through deep neural networks, improving the accuracy and efficiency of speech recognition.

[0050] Furthermore, in some preferred embodiments, the above-mentioned speech recognition results can also be optimized, processed and extracted through the question-answering model in the computing platform 120, such as Deepseek, Chatgpt, etc.

[0051] Specifically, if Figure 3 As shown, the vehicle system 110 may also include a text processing and optimization module, which can be used to receive the speech recognition results transmitted by the speech recognition module, and call the question-answering model in the cloud server platform 121 or the local computing platform 122 to perform text adjustment optimization processing on the original text generated by the speech recognition results, so as to obtain an initial text to be projected that is coherent and in line with the original intention of the people in the car.

[0052] In some optional embodiments, text adjustment may include spelling check, grammar check, context correction, etc. Text adjustment through spelling check can be used to check whether the spelling of words in the recognition results is correct and to correct misspelled words. Text adjustment through grammar check can be used to check whether the grammar of the recognition results is correct and to adjust ungrammatical sentences. Text adjustment through context correction can be used to further correct the recognition results based on context information, for example, judging the correct form of a word based on the context, deleting redundant content, etc. By optimizing the text adjustment of the original text, an initial text to be projected that is close to and consistent with the original intention of the driver and passengers can be generated.

[0053] Continue as Figure 2 As shown, the above-mentioned vehicle light projection control method provided by the present invention may further include step S220: determining whether the initial text includes preset standard information.

[0054] Specifically, combined Figure 3 It is understood that in some embodiments, the text processing and optimization module in the vehicle-mounted system 110 can determine whether the initial text includes preset standard information. The preset standard information may include projection keywords and regulatory pattern names. Projection keywords may include words with obvious projection direction, such as "text display." The regulatory pattern name may refer to a standard pattern specified in regulations for headlight projection. Regulations for headlight projection require that the projection pattern be unambiguous.

[0055] like Figure 2 As shown, the above-mentioned vehicle light projection control method provided by the present invention may further include step S230: in response to the initial text including the projection keyword, generating a first projection pattern of the guidance object based on the projection keyword through the text-based graph model.

[0056] Specifically, in some optional embodiments, if the initial text includes such projection keywords, such as "text display" or similar words, the first projection text corresponding to the guidance object can be extracted from the initial text according to the guidance of the projection keyword.

[0057] For example, if the original text generated by speech recognition is "Display text Ha hello ah," the question-answering model can optimize the original text by removing duplicate content and adjusting the text to return a coherent initial text "Display text hello ah" that matches the occupant's original intention. Because the initial text includes the projection keyword "display text," the "hello ah" following "display text" is the target of its guidance, and therefore "hello ah" is extracted as the first projected text.

[0058] Furthermore, preferably, the first projection text can also be replaced by multiple different language versions with the same meaning. For example, the extracted first projection text can also be replaced by the English word "HELLO".

[0059] Continue as Figure 3 As shown, the vehicle system 110 may also include a prompt word generation module, which can supplement the projected text and further generate a projection pattern that meets the requirements of the headlight projection. Normally, the headlights can project patterns of pictures in any format, such as jpg, gif, png, bmp, etc., but all pictures are projected in black and white, and the pixels are 10,000 to 20,000. Here, "meeting the requirements of headlight projection" can be understood as content that is more suitable for headlight projection. Generally speaking, pictures suitable for headlight projection can preferably include the following features: obvious black and white contrast, centered content, and simple strokes. Such pictures are clearer and more distinct when projected through the headlights. Therefore, the prompt word generation module can be used to assist in generating the above-mentioned style of pattern.

[0060] Specifically, in this embodiment, the prompt word generation module can obtain the first projected text and, based on it, generate a first prompt word that meets the requirements of vehicle headlight projection to supplement the first projected text. For example, the first prompt word can include "white text on a black background." Subsequently, the text image generation module can combine the first projected text and the first prompt word and invoke the text image model of the cloud server platform 121 or the local computing platform 122 to generate the first projection pattern.

[0061] For example, continuing with the above example, assuming that the first projected text returned is the English word "HELLO", the English word can be supplemented through the prompt word generation module. On the basis of "HELLO", add "black background picture", "centered text", "white font" and other words to form a complete first prompt word, and these contents are given to the text generation model for generation through the text generation module. The text generation model can eventually generate a first projection pattern with white text on a black background and the English word "HELLO" centered, and return it to the vehicle system 110.

[0062] Continue as Figure 2 As shown, the above-mentioned vehicle light projection control method provided by the present invention may further include step S240: in response to the initial text not including the preset standard information, generating a second projection pattern based on the initial text through the text-based image model.

[0063] Specifically, combined Figure 3It is commonly understood that in some embodiments, the text processing and optimization module in the vehicle system 110 determines that the initial text does not include the preset standard information of the above-mentioned projection keywords or regulatory pattern names. At this time, the initial text can be used as the second projection text.

[0064] For example, if the original text generated by speech recognition is "Two girls sitting on the grass playing with water," the Q&A model can optimize the original text by removing redundant content and adjusting the text to return a coherent initial text to be projected, "Two girls sitting on the grass playing with water," which aligns with the occupants' intended meaning. Because the initial text contains neither projection keywords (such as "display text") nor the name of the regulatory pattern, "Two girls sitting on the grass playing with water" is used as the second projection text, and the Text-Image model is used to generate a corresponding projection pattern.

[0065] like Figure 3 As shown, in this embodiment, the prompt word generation module can obtain the second projected text and, based on it, generate a second prompt word that meets the requirements of the headlight projection to supplement the second projected text. Subsequently, the text image generation module can combine the second projected text and the second prompt word and call the text image model of the cloud server platform 121 or the local computing platform 122 to generate the second projection pattern.

[0066] For example, continuing with the above example, assuming that the second projected text returned is "Two girls sitting on the grass playing with water", this content can be supplemented by the prompt word generation module. On the basis of "Two girls sitting on the grass playing with water", add "the main body is white", "simple drawing style", "do not show specific details", "show outline and fill with white", and "black background" and other words to form a complete first prompt word, and these contents are given to the text-generated image model for image generation through the text-generated image block module. The text-generated image model can eventually generate a pattern with a black background and white content. The content displayed by the pattern is the content described by the driver and passengers, and it is returned to the vehicle system 110.

[0067] Through the above-mentioned embodiments including the first projection pattern and the second projection pattern, the present invention can understand the intentions of the driver and passengers more clearly and accurately through the question-and-answer model, thereby improving the user experience. The Wenshengtu AI large model can break through the limitations of the car light projection content, understand the content described by the driver and passengers, and create freely, thereby giving the car lights more flexibility and playability, and realizing a more diversified, convenient and creative user experience of car light projection, allowing users to freely create projection content according to personal preferences, and also increasing user interactivity and participation.

[0068] Further, continue as Figure 3As shown, in some other optional embodiments, some traffic-related regulation patterns may also be pre-stored in the control system of the present invention, and they are named according to their meanings. For example, certain patterns may be named "Collision Warning", "Zebra Crossing", "Skid Warning", etc., and are pre-stored in the regulation pattern library in the control system.

[0069] If the initial text includes the names of such regulation patterns, the name of the regulation pattern can be used as the third projection text, and then the pre-stored pattern corresponding to the name of the regulation pattern can be directly called from the regulation pattern library as the third projection pattern. As Figure 3 shown, these regulation patterns can directly read the pre-stored local pictures for projection without passing through the prompt word generation module, the text-to-image generation module, and the text-to-image large model.

[0070] For example, when the original text generated by the voice recognition result is "Project a skid warning sign", through the question and answer large model to optimize the text adjustment to complete the sentence meaning of the original text, the initial text to be projected that is smooth and in line with the intention of the vehicle occupants can be returned, "Project a skid warning sign". Since the initial text includes the word "skid warning" which is the name of a regulation pattern, therefore, "skid warning" can be directly used as the third projection text.

[0071] Continue as Figure 3 shown, by searching the regulation pattern library in the vehicle-mounted system 110, the pre-stored pattern corresponding to the third projection text can be directly called as the third projection pattern without supplementing the third projection text.

[0072] After that, continue as Figure 2 shown, the above control method for headlight projection provided by the present invention may further include step S250: at least post-process the first projection pattern and / or the second projection pattern so that the first projection pattern and / or the second projection pattern are projected and displayed without distortion according to the headlight pixels.

[0073] Specifically, in combination with Figure 3 it can be understood that in some embodiments, the vehicle-mounted system 110 may further include a picture post-processing module, which can be used to post-process the first projection pattern and / or the second projection pattern and / or the third projection pattern. For example, it may include links such as picture size adjustment, trapezoidal calibration, grayscale conversion, pixel parsing, data sending, etc.

[0074] Specifically, in some optional embodiments, the RGB three-channel maps of the first projection pattern and / or the second projection pattern and / or the third projection pattern may be pre-processed by grayscale conversion to obtain the corresponding grayscale single-channel maps.

[0075] Since normal images have three RGB color channels, and current car headlights include two on and off states, black and white, as well as brightness, therefore, optionally, in this embodiment, the RGB three-channel image can be converted into a single-channel grayscale image, so that the color image of the first projection pattern and / or the second projection pattern and / or the third projection pattern (i.e., the RGB three-channel image) can be projected and displayed via the car lights.

[0076] Those skilled in the art will understand that the above-mentioned step of grayscale preprocessing of the RGB three-channel image is only a non-limiting embodiment provided by the present invention, which is intended to clearly demonstrate the main concept of the present invention and provide a specific solution that is convenient for the public to implement, rather than to limit the scope of protection of the present invention. Optionally, in other embodiments, if future automobile lights eliminate the two black and white on and off states and the brightness limit, those skilled in the art may also omit this grayscale step based on the concept of the present invention and directly proceed to the subsequent steps to achieve the same technical effect.

[0077] Afterwards, please combine Figure 4 Common understanding, Figure 4 A schematic diagram of adjusting the size of a projection pattern according to some embodiments of the present invention is shown.

[0078] like Figure 4 As shown, the image size of the corresponding grayscale single-channel image can be adjusted according to the pixel size of the headlight light source to obtain the corresponding original pre-projection pattern.

[0079] Combine Figure 3 and Figure 4 ,like Figure 4 Taking the first projection pattern shown as an example, assuming the pixels of the Micro-LED headlight light source are 64*256 or 80*320, the image post-processing module can first execute step S410 to obtain the second projection pattern. Then, it can execute step S420 to obtain the minimum bounding rectangle corresponding to the first projection pattern.

[0080] Specifically, in this embodiment, the above steps have converted the image into a grayscale single-channel image, where a pixel value of 0 represents black. Therefore, non-black pixels (i.e., non-0 pixels) need to be projected. Therefore, by traversing each pixel, the coordinates corresponding to non-0 pixel values can be found, and their maximum and minimum values in the horizontal and vertical directions can be calculated, thereby obtaining the minimum bounding rectangle of the content to be projected.

[0081] Afterwards, step S430 can be executed to crop the minimum circumscribed rectangle. Finally, step S440 can be executed to scale the image to the same size as the light source pixels. For example, the rectangle can be cropped and scaled to 64*256 or 80*320. Through the above steps S410 to S440, the image can be adjusted to the size of the corresponding pixels of the headlights according to the pixel size of the headlight light source installed on the vehicle, and the original pre-projection pattern after the image size is adjusted can be obtained so that the first projection pattern and / or the second projection pattern / or the third projection pattern can be projected and displayed via the corresponding LEDs in the headlight module.

[0082] Furthermore, because headlights typically have a downward tilt, this can cause trapezoidal distortion when projecting a rectangular image onto the ground. That is, the projected image is distorted from a rectangle to a trapezoid. To address this, in some preferred embodiments, an image post-processing module can perform trapezoidal correction on the original pre-projected pattern projected onto the ground to obtain a corrected, official pre-projected pattern.

[0083] Specifically, it can be combined Figure 5 Common understanding, Figure 5 A schematic diagram of performing trapezoidal correction on an original pre-projection pattern according to some embodiments of the present invention is shown.

[0084] like Figure 5 As shown, in some embodiments, step S510 can be first performed to input the original pre-projected pattern into the image post-processing module and measure its four original corner coordinates, (org1_x, org1_y), (org2_x, org2_y), (org3_x, org3_y), and (org4_x, org4_y). Then, step S520 can be performed to project the original pre-projected pattern onto the ground and measure its four projection corner coordinates, (proj1_x, proj1_y), (proj2_x, proj2_y), (proj3_x, proj3_y), and (proj4_x, proj4_y). Then, step S530 can be performed to calculate the perspective transformation matrix based on the original corner coordinates and the projection corner coordinates. Thereafter, step S540 can be performed to invert the perspective transformation matrix to obtain an inverse perspective transformation matrix. Finally, step S550 may be executed to perform an inverse perspective transformation on the original pre-projection pattern using an inverse perspective transformation matrix to obtain a reverse compensation image of the original pre-projection pattern as a formal pre-projection pattern.

[0085] Specifically, the inverse perspective transformation matrix can be multiplied by the original pre-projection pattern to perform an inverse perspective transformation on the original pre-projection pattern, thereby obtaining a reverse compensation image of the original pre-projection pattern as the formal pre-projection pattern. The corrected formal pre-projection pattern, when projected onto the ground, can appear as an undistorted rectangle.

[0086] Furthermore, in some optional embodiments, after receiving the formal pre-projection pattern sent via the image post-processing module, the headlight electronic control unit 130 in the vehicle system 110 can also obtain the pixel value of each pixel in the grayscale single-channel image corresponding to the formal pre-projection pattern.

[0087] Optionally, each pixel in the grayscale single-channel image can correspond to one or more LEDs in the headlight module, and the value of each pixel ranges from 0 to 255, where 0 represents that the LED is off, 255 represents that the LED is brightest, x / 255 represents the degree of lighting, and x represents the pixel value in the grayscale single-channel image. Therefore, after the headlight electronic control unit 130 obtains the pixel value x (0≤x≤255) of each pixel in the grayscale single-channel image corresponding to the formal pre-projection pattern, it can control the brightness and on / off state of each corresponding LED in the headlight module 140 through the pixel value of each pixel. Then, the headlight electronic control unit 130 can control the lighting of the corresponding LED light sources in the headlight module 140 after the brightness adjustment is completed, and project the formal pre-projection pattern onto the ground without distortion according to the headlight pixels.

[0088] Through the above-mentioned multiple embodiments, the present invention inputs a voice message to the vehicle-mounted system 110, which then calls a large model algorithm on a cloud-based or local high-performance computing platform 120 to generate, call, and post-process the projection pattern. Ultimately, the projection data is sent to the vehicle-mounted module via the vehicle-mounted ECU for projection. Compared to existing technologies, the present invention allows the user to describe the image to be projected through voice, create, and project the image, making the use of vehicle-mounted projection more flexible and convenient for drivers and passengers.

[0089] Those skilled in the art will understand that the above-mentioned vehicle system 110, which includes multiple functional modules, a voice recognition module, a text processing and optimization module, and other independent solutions, is only a non-limiting implementation method provided by the present invention, which is intended to clearly demonstrate the main concept of the present invention and provide a specific solution that is easy for the public to implement, rather than to limit the scope of protection of the present invention. Optionally, in other embodiments, those skilled in the art may also integrate some or all of these functional modules into one or several control modules based on the concept of the present invention, so that each of the above-mentioned independent functional modules is not limited to performing the respective functional roles described above, but can also perform the functions of some other functional modules. For example, the text processing and optimization module and the prompt word generation module can be integrated into a total module for text optimization so that it can perform the two functions of the two independent modules.

[0090] Those skilled in the art will further appreciate that the steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read and write information from / to the storage medium. In an alternative, the storage medium may be integrated into the processor. The processor and storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and storage medium may reside in a user terminal as discrete components.

[0091] In one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functions may be stored on or transmitted via a computer-readable medium as one or more instructions or codes. Computer-readable media include both computer storage media and communication media, including any media that facilitates the transfer of a computer program from one location to another. A storage medium may be any available medium that can be accessed by a computer. By way of example and not limitation, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Any connection is also properly referred to as a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwaves are included in the definition of medium. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk and Blu-ray disc, where disks typically reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.

[0092] Although the above methods are illustrated and described as a series of acts for simplicity of explanation, it is to be understood and appreciated that these methods are not limited by the order of the acts, as some acts may occur in a different order and / or concurrently with other acts from those illustrated and described herein or not illustrated and described herein but understandable to those skilled in the art according to one or more embodiments.

[0093] In summary, the present invention provides a control method for car light projection, a control system for car light projection, and a computer-readable storage medium, which not only make the car light projection process simpler and faster, but also can create personalized car light projection patterns according to the user's imagination, breaking through the limitations of pre-made content of car light projection, enhancing user experience, and improving the interactivity, diversity and playability of car light projection content.

[0094] The previous description of the disclosure is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the disclosure is not intended to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for controlling vehicle light projection, characterized in that: The following steps are involved: According to the voice input, the initial text to be projected is obtained; Determining whether the initial text includes preset standard information, wherein the preset standard information includes projection keywords and regulatory pattern names; In response to the initial text including a projection keyword, generating a first projection pattern of a guidance object based on the projection keyword through a text-based graph model; In response to the initial text not including the preset standard information, generating a second projection pattern based on the initial text by using the Wensheng graph model; and At least the first projection pattern and / or the second projection pattern are post-processed so that the first projection pattern and / or the second projection pattern are projected and displayed without distortion according to the pixels of the vehicle light.

2. The control method according to claim 1, wherein: In response to the initial text including a projection keyword, the step of generating a first projection pattern of a guidance object based on the projection keyword by using the cultural graph model includes: In response to the initial text including the projection keyword, extracting a first projection text corresponding to the guidance object from the initial text according to the guidance of the projection keyword; Based on the first projection text, generating a first prompt word that meets the requirements of vehicle light projection; and The first projection pattern is generated by combining the first projection text and the first prompt word through the large model of the cultural image.

3. The control method according to claim 1, wherein: In response to the initial text not including the preset standard information, the step of generating a second projection pattern based on the initial text by using a large model of a text map comprises: In response to the initial text not including the preset standard information, using the initial text as the second projected text; Based on the second projection text, generating a second prompt word that meets the requirements of the headlight projection; and The second projection text and the second prompt word are combined and the second projection pattern is generated through the large model of the cultural image.

4. The control method according to claim 1, wherein: The step of determining whether the initial text includes preset standard information further includes: In response to the initial text including the regulatory pattern name, using the regulatory pattern name as the third projected text; and A pre-stored pattern corresponding to the third projection text is called from a regulatory image library as the third projection pattern.

5. The control method according to claim 4, wherein: The step of post-processing at least the first projection pattern and / or the second projection pattern so that the first projection pattern and / or the second projection pattern are projected and displayed without distortion according to the vehicle light pixels comprises: Performing grayscale preprocessing on the RGB three-channel image of the first projection pattern and / or the second projection pattern / or the third projection pattern to obtain a corresponding grayscale single-channel image; Adjusting the image size of the corresponding grayscale single-channel image according to the pixel size of the headlight light source to obtain the corresponding original pre-projection pattern; Performing trapezoidal correction on the original pre-projection pattern projected onto the ground to obtain a corrected formal pre-projection pattern; and The formal pre-projection pattern is transmitted to the vehicle light module so as to be projected and displayed via the vehicle light module.

6. The control method according to claim 5, wherein: The step of performing trapezoidal correction on the original pre-projection pattern projected onto the ground to obtain a corrected formal pre-projection pattern comprises: Obtaining four original angular coordinates of the original pre-projection pattern and four projection angular coordinates projected onto the ground; Calculating a perspective transformation matrix based on the original angular coordinates and the projection angular coordinates; Inverting the perspective transformation matrix to obtain an inverse perspective transformation matrix; and The original pre-projection pattern is subjected to an inverse perspective transformation through the inverse perspective transformation matrix to obtain an inverse compensation image of the original pre-projection pattern as the formal pre-projection pattern.

7. The control method according to claim 5, wherein: The step of transmitting the formal pre-projection pattern to the vehicle light module for projection display via the vehicle light module includes: Obtaining a pixel value of each pixel in the grayscale single-channel image corresponding to the formal pre-projection pattern; Controlling the brightness of corresponding LEDs in the vehicle light module based on the pixel values of the pixels; and Light up each LED in the vehicle light module after brightness adjustment, and project the formal pre-projection pattern onto the ground for display.

8. The control method according to claim 1, wherein: The step of obtaining the initial text to be projected according to the voice input includes: Acquiring voice input information from a person in the vehicle, wherein the voice input information at least includes vehicle light projection content; Performing voice recognition on the voice input information to obtain a voice recognition result; and The speech recognition result is optimized through the question-answering model to obtain the initial text to be projected.

9. The control method according to claim 8, wherein: The step of optimizing the speech recognition result by using the large question-answer model to obtain the initial text to be projected includes: The original text generated by the speech recognition result is subjected to text adjustment to obtain the initial text to be projected that is coherent and in accordance with the original intention of the occupant of the vehicle.

10. A control system for vehicle light projection, characterized in that: include: The vehicle computer system is used to obtain the initial text to be projected based on the voice input; Determining whether the initial text includes preset standard information, wherein the preset standard information includes projection keywords and regulatory pattern names; a computing platform, communicatively connected to the vehicle-mounted system, comprising a large Wensheng diagram model, configured to include a projection keyword in the initial text and generate, via the large Wensheng diagram model, a first projection pattern of a guidance object based on the projection keyword; and, when the initial text does not include the preset standard information, generate, via the large Wensheng diagram model, a second projection pattern based on the initial text; and return the second projection pattern to the vehicle-mounted system, so that the vehicle-mounted system performs post-processing on at least the first projection pattern and / or the second projection image. a headlight electronic control unit, communicatively connected to the vehicle computer system, and configured to at least obtain the first projection pattern and / or the second projection image after post-processing; and The headlight module is communicatively connected to the headlight electronic control unit and is used to project and display the first projection pattern and / or the second projection image after post-processing without distortion according to the headlight pixels.

11. The control system according to claim 10, wherein: The vehicle system includes: a text processing and optimization module, configured to determine whether the initial text includes preset standard information; in response to the initial text not including the preset standard information, use the initial text as the first projected text; in response to the initial text including the projected keyword, extract a second projected text corresponding to the guidance object from the initial text according to the guidance of the projected keyword; and The regulatory image library module is used to determine in the text processing and optimization module that the initial text includes the regulatory pattern name, and use the regulatory pattern name as the third projection text; and call the pre-stored pattern corresponding to the third projection text from the regulatory image library as the third projection pattern.

12. The control system according to claim 11, wherein: The vehicle system includes: a prompt word generation module, configured to generate, based on the first projection text and / or the second projection text, corresponding first prompt words and / or second prompt words that meet the requirements of vehicle light projection; and The text-generated image module is configured to combine the first projected text and the first prompt word and / or combine the second projected text and the second prompt word, call the text-generated image model, and generate the first projection pattern and / or the second projection pattern.

13. The control system according to claim 11, wherein: The vehicle system includes: The image post-processing module is used to perform grayscale pre-processing on the RGB three-channel image of the first projection pattern and / or the second projection pattern when post-processing at least the first projection pattern and / or the second projection pattern, so as to obtain a corresponding grayscale single-channel image; adjust the image size of the corresponding grayscale single-channel image according to the pixel size of the headlight light source, so as to obtain a corresponding original pre-projection pattern; perform trapezoidal correction on the original pre-projection pattern projected onto the ground, so as to obtain a corrected formal pre-projection pattern; and send the formal pre-projection pattern to the headlight electronic control unit so as to control the projection display of the headlight module via the headlight electronic control unit.

14. The control system according to claim 13, wherein: The vehicle light electronic control unit is further configured to, after obtaining the formal pre-projection pattern, obtain a pixel value of each pixel in a grayscale single-channel image corresponding to the formal pre-projection pattern; and control the brightness of each corresponding LED in the vehicle light module based on the pixel value of each pixel; and The vehicle light module is further used to light up the LEDs after brightness adjustment is completed, and project the formal pre-projection pattern onto the ground for display.

15. The control system according to claim 11, wherein: The vehicle system also includes a voice input module and a voice recognition module, and the computing platform also includes a voice recognition algorithm and a question-answering model, wherein: The voice input module is used to obtain voice input information from people in the car, wherein the voice input information includes the content projected by the car lights; The speech recognition module is used to receive the speech input information and parse it into corresponding coded data; The speech recognition algorithm is used to perform speech recognition based on the encoded data to obtain a speech recognition result; and The large question-answer model is used to optimize the speech recognition result to obtain the initial text to be projected.

16. The control system according to claim 15, wherein: The computing platform includes a cloud server platform or a local computing platform, wherein: The cloud server platform is connected to a plurality of vehicle systems, and is used to enable the plurality of vehicle systems to share the large model of the text graph, the speech recognition algorithm and the large model of question and answer; and The local computing platform is connected to a single vehicle system, and is used to enable the text-graph large model, the speech recognition algorithm and the question-answering large model to serve the single vehicle system in a targeted manner.

17. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by the processor, the vehicle light projection control method according to any one of claims 1 to 9 is implemented.

Images