Vehicle display system detection method and device, electronic equipment and storage medium

Abstract

Embodiments of the present application provide a vehicle-mounted display system detection method and device, electronic equipment and storage medium, relating to the field of vehicle-mounted screen detection, the method comprising: collecting an output image of a vehicle-mounted display system, preprocessing the output image to obtain an image to be analyzed, inputting the image to be analyzed into a pre-trained deep learning model, and outputting a detection result of the display state of the vehicle-mounted display system, so as to improve the detection accuracy of the vehicle-mounted display system.

Description

Technical Field

[0001] This invention relates to the field of vehicle screen testing, and more specifically, to a method, apparatus, electronic device, and storage medium for testing vehicle display systems. Background Technology

[0002] With the rapid development of intelligent connected vehicles, the functions of in-vehicle infotainment systems are becoming increasingly complex. As the core carrier of human-machine interaction, the display interface directly affects user experience and driving safety. Therefore, quality testing of in-vehicle displays has become a crucial step in research and development, production, and quality control.

[0003] Currently, the testing of in-vehicle display systems mainly relies on manual methods, with common approaches including screen screenshot analysis and external camera photography.

[0004] Screenshot detection method: This method obtains the digital image of the currently displayed screen through the built-in debugging interface or software function of the vehicle display system.

[0005] Although it is easy to operate and unaffected by ambient light, it has obvious limitations. It can only obtain logical "output signals" and cannot reflect the actual physical display effect. It is difficult to detect visually visible defects such as local dead spots, uneven brightness, abnormal contrast, and color cast. It is prone to missing detections when the GPU rendering is normal but the display hardware is abnormal (such as backlight failure or LCD panel damage).

[0006] External camera shooting method: Use an external camera to take pictures or record videos of the in-vehicle display system screen, and then analyze the display status through image processing algorithms.

[0007] Although this method can simulate the perspective of a real user, it still faces many challenges. It is susceptible to interference from ambient light, with severe reflections in strong light and low signal-to-noise ratios in low light, leading to image distortion. Deviations in shooting angle can cause perspective distortion, affecting the accuracy of subsequent image analysis. The equipment layout is complex, making it difficult to achieve automated pipeline integration. Image acquisition stability is poor, repeatability is not good, and it is not conducive to long-term consistency testing. Summary of the Invention

[0008] The purpose of this invention is to provide a method, apparatus, electronic device, and storage medium for testing in-vehicle display systems, which can improve the accuracy of testing in-vehicle display devices.

[0009] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: In a first aspect, embodiments of this application provide a method for detecting an in-vehicle display system, the method comprising: Acquire images output from the vehicle-mounted display system; The output image is preprocessed to obtain the image to be analyzed; The image to be analyzed is input into a pre-trained deep learning model, which outputs the detection result of the display status of the vehicle display system.

[0010] In an optional implementation, the step of preprocessing the output image to obtain the image to be analyzed includes: The output image is subjected to at least one of the following: denoising, contrast enhancement, grayscale conversion, or multi-channel feature extraction, wherein the denoising is performed using a Gaussian filtering algorithm, and the contrast enhancement is performed using a histogram equalization method.

[0011] In an optional implementation, the step of inputting the image to be analyzed into a pre-trained deep learning model and outputting the detection result of the display status of the vehicle display system includes: Determine the average brightness of the image to be analyzed; The average brightness is compared with an adaptive threshold; When the average brightness is greater than or equal to the adaptive threshold, the display state of the vehicle display system is determined to be on. When the average brightness is less than the adaptive threshold, the display status of the vehicle display system is determined to be not normally lit.

[0012] In an optional implementation, the method further includes: Determine the average brightness and brightness standard deviation of the image to be analyzed; Calculate the product of the brightness standard deviation and the adjustment coefficient, wherein the adjustment coefficient is used to control the sensitivity of the threshold to the brightness standard deviation; The sum of the average brightness and its product is calculated as an adaptive threshold.

[0013] In an optional implementation, the method further includes: When the display state of the vehicle display system is not normally lit, anomaly detection is performed on the display state of the vehicle display system. The anomaly detection includes at least one of the following: gray screen detection, screen distortion detection, screen freeze detection, color cast detection, and no signal detection.

[0014] In an optional implementation, the step of detecting anomalies in the display status of the vehicle-mounted display system includes: Determine the standard deviation of the brightness of the image to be analyzed; When the brightness standard deviation is lower than the standard deviation threshold, it is determined to be a gray screen; The high-frequency energy density of the image to be analyzed is analyzed by frequency domain transformation; When the high-frequency energy density is higher than the standard high-frequency energy density, it is determined to be a screen distortion. Determine the differences between multiple consecutive frames of the image to be analyzed; When the difference approaches zero, it is determined to be a screen freeze; Calculate the mean values ​​of the RGB three channels of the image to be analyzed; The proportion of each channel is calculated based on the average of the three RGB channels; If the ratio of any channel exceeds the normal range, it is judged as color cast; OCR technology is used to identify whether the image to be analyzed contains keywords without a signal. If so, it is determined to be a no-signal state.

[0015] In an optional implementation, the method further includes: Load the pre-trained MobileNetV2 model as the base network; A first basic network is obtained by adding a fully connected layer and a Dropout layer after the basic network. Freeze the convolutional layers of the first base network to obtain the second base network; The second basic network is trained to obtain the first trained model; The convolutional layers of the first model after training are unfrozen to obtain the third base network; The third basic network is trained to obtain a pre-trained deep learning model.

[0016] Secondly, embodiments of this application provide a vehicle-mounted display system testing device, the device comprising: The acquisition module is used to acquire images output by the vehicle-mounted display system; The processing module is used to preprocess the output image to obtain the image to be analyzed; input the image to be analyzed into a pre-trained deep learning model, and output the detection result of the display status of the vehicle display system.

[0017] Thirdly, embodiments of this application provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the vehicle display system detection method.

[0018] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the vehicle display system detection method.

[0019] This application has the following beneficial effects: This application collects images output from an in-vehicle display system, preprocesses the output images to obtain images to be analyzed, and inputs the images to be analyzed into a pre-trained deep learning model to output the detection results of the display status of the in-vehicle display system, thereby improving the detection accuracy of the in-vehicle display system. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A block diagram of an electronic device provided in an embodiment of the present invention; Figure 2 This is one of the flowcharts illustrating a vehicle-mounted display system testing method provided in an embodiment of the present invention; Figure 3 This is a second flowchart illustrating a method for detecting an in-vehicle display system according to an embodiment of the present invention. Figure 4 This is a third flowchart illustrating a method for detecting an in-vehicle display system according to an embodiment of the present invention. Figure 5 The fourth flowchart illustrates a method for detecting an in-vehicle display system according to an embodiment of the present invention. Figure 6 The fifth flowchart illustrates a method for detecting an in-vehicle display system according to an embodiment of the present invention. Figure 7 This is a structural block diagram of a vehicle display system testing device provided in an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0024] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0025] In the description of this invention, it should be noted that if terms such as "upper," "lower," "inner," or "outer" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of this invention is usually placed, they are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0026] Furthermore, the terms "first" and "second" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0027] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "set up," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0028] Extensive research has revealed that traditional vehicle-mounted display system detection methods primarily rely on manual photography or screenshots. While screenshots can acquire digital images, they cannot simulate the actual viewing experience for a real user, potentially leading to missed dead pixels and uneven brightness. External camera photography is susceptible to environmental lighting, reflections, and angular deviations, resulting in failed images or misjudgments.

[0029] In view of the above-mentioned problems, this embodiment provides a method, device, electronic device and storage medium for detecting vehicle display systems. It can acquire the output image of the vehicle display system, preprocess the output image to obtain the image to be analyzed, input the image to be analyzed into a pre-trained deep learning model, and output the detection result of the display status of the vehicle display system, so as to improve the detection accuracy of the vehicle display system. The solution provided in this embodiment will be described in detail below.

[0030] This embodiment provides an electronic device capable of detecting in-vehicle display systems. In one possible implementation, the electronic device can be a user terminal, such as, but not limited to, a server, smartphone, personal computer (PC), tablet computer, personal digital assistant (PDA), mobile internet device (MID), and in-vehicle display system.

[0031] Please refer to Figure 1 , Figure 1 This is a schematic diagram of the structure of the electronic device 100 provided in the embodiments of this application. The electronic device 100 may further include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown. Figure 1 The components shown can be implemented using hardware, software, or a combination thereof.

[0032] The electronic device 100 includes an in-vehicle display system detection device 110, a memory 120, and a processor 130.

[0033] The components of the memory 120 and processor 130 are electrically connected directly or indirectly to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines. The vehicle display system detection device 110 includes at least one software function module that can be stored in the memory 120 in the form of software or firmware or embedded in the operating system (OS) of the electronic device 100. The processor 130 is used to execute the executable modules stored in the memory 120, such as the software function modules and computer programs included in the vehicle display system detection device 110.

[0034] The memory 120 may be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory 120 is used to store programs, and the processor 130 executes the programs after receiving execution instructions.

[0035] Please refer to Figure 2 , Figure 2 For application Figure 1 The flowchart below shows a method for testing an in-vehicle display system of an electronic device 100. The method includes a detailed description of each step.

[0036] S201: Acquire images output by the vehicle display system.

[0037] S202: Preprocess the output image to obtain the image to be analyzed.

[0038] S203: Input the image to be analyzed into a pre-trained deep learning model and output the detection results of the display status of the vehicle display system.

[0039] A video capture card can be used, with one end connected to the vehicle display system and the other end connected to electronic devices via a DP to HDMI adapter cable. Uncompressed raw frame data, i.e., the output image of the vehicle display system, can be directly obtained from the vehicle display system through OpenCV, thereby completely avoiding environmental noise, geometric distortion and color deviation introduced by optical acquisition.

[0040] Video capture cards are prone to introducing high-frequency salt-and-pepper noise, edge ringing artifacts, and chromatic crosstalk in DP→HDMI signal conversion, USB bandwidth limitations, and driver compatibility issues. To unify display differences across vehicle models: different vehicle UIs use different brightness benchmarks, such as lower overall grayscale in dark mode, lower contrast settings, and different Gamma curves (sRGB vs DCI-P3), requiring normalization to eliminate device dependencies. To adapt to lightweight model input requirements: MobileNetV2 fine-tuning models have clear constraints on input size (224×224), pixel distribution ([0,1] interval), and signal-to-noise ratio (SNR ≥ 25dB), and preprocessing must ensure that the input data meets the inference stability threshold.

[0041] Preprocessing may include at least one of the following: denoising, contrast enhancement, and grayscale conversion or multi-channel feature extraction of the output image. The denoising process uses a Gaussian filtering algorithm, and the contrast enhancement process uses a histogram equalization method to obtain the image to be analyzed.

[0042] In one example, the output image captured by the video capture card is converted to grayscale, then denoised using Gaussian blur, and finally normalized to obtain the image to be analyzed. This image is then used as input to a pre-trained deep learning model, which outputs the detection results of the display status of the vehicle display system.

[0043] The display results may include whether the screen is on, gray screen, distorted screen, frozen screen, color cast, and no signal.

[0044] Compared with existing technologies, this application not only solves the core pain points of traditional methods such as large environmental interference, high detection error, and inability to achieve real-time detection, but also realizes intelligent identification and classification output of various display anomalies, which has outstanding practical value and technological advancement.

[0045] There are several ways to input the image to be analyzed into a pre-trained deep learning model and output the detection results of the display status of the vehicle display system. In one implementation method, such as... Figure 3 As shown, it includes the following steps: S301: Determine the average brightness of the image to be analyzed.

[0046] S302: Compare the average brightness with the adaptive threshold.

[0047] S303: When the average brightness is greater than or equal to the adaptive threshold, the display state of the vehicle display system is determined to be on.

[0048] S304: When the average brightness is less than the adaptive threshold, the display status of the vehicle display system is determined to be not normally lit.

[0049] In digital image processing, the average brightness of an image refers to the arithmetic mean of the brightness values ​​of all pixels in the image, which reflects the overall brightness of the image being analyzed.

[0050] Since the image to be analyzed is a preprocessed image, and the preprocessing method includes grayscale conversion, when determining the average brightness of the image to be analyzed, the pixel value of each pixel in the image to be analyzed can be determined, and the average value of each pixel value can be calculated as the average brightness of the image to be analyzed.

[0051] The average brightness of the image to be analyzed is compared with an adaptive threshold. If the average brightness is greater than or equal to the adaptive threshold, the vehicle display system is determined to be on normally and the display status is on. If the average brightness is less than the adaptive threshold, the vehicle display system is determined to be on but not on, and the display status is not on normally. A prompt message indicating that the display system is not on normally is also output.

[0052] There are multiple ways to determine the adaptive threshold. In one implementation, such as... Figure 4 As shown, it includes the following steps: S401: Determine the average brightness and brightness standard deviation of the image to be analyzed.

[0053] S402: Calculate the product of the standard deviation of brightness and the adjustment coefficient.

[0054] The adjustment coefficient is used to control the sensitivity of the threshold to the standard deviation of brightness.

[0055] S403: Calculate the sum of the average brightness and its product, and use it as an adaptive threshold.

[0056] Traditional image detection methods typically use a fixed brightness threshold to determine whether a device's screen is on. However, due to significant differences in factory brightness settings, backlight types, and user interface themes among different automotive display systems, a single fixed threshold is difficult to adapt to various vehicle models and usage scenarios, easily leading to misjudgments or missed detections.

[0057] This invention employs an adaptive threshold brightness determination mechanism. Specifically, it calculates the average brightness and brightness standard deviation of the image to be analyzed in real time, and dynamically generates the adaptive threshold used for this detection based on these values. The expression for this threshold is as follows: ; in, Average brightness The standard deviation of brightness, This is the adjustment coefficient.

[0058] When the average brightness of the image to be analyzed is less than the adaptive threshold, it is determined to be either not lit or has abnormal brightness.

[0059] The adaptive threshold is determined based on the average brightness and the standard deviation of brightness. The screen brightness judgment is based on the adaptive threshold. The advantage of this mechanism is that for images that are generally dark but have clear content, such as dark UI, the standard deviation is large, so the adaptive threshold is increased accordingly to avoid misjudgment. However, for black screen or gray screen images with no display content, the average brightness and the standard deviation of brightness are extremely low, resulting in a very small adaptive threshold, which can still be accurately identified.

[0060] When the in-vehicle display system is not normally lit, an anomaly detection is performed on the display status. This anomaly detection includes at least one of the following: gray screen detection, screen flickering detection, screen freezing detection, color cast detection, and no signal detection. When performing gray screen detection on the in-vehicle display system, if... Figure 5 As shown, it includes the following steps: S501: Determine the standard deviation of brightness of the image to be analyzed.

[0061] S502: When the brightness standard deviation is lower than the preset brightness standard deviation, it is judged as a gray screen.

[0062] After converting the output image acquired by the video capture card to grayscale, the grayscale value distribution of all pixels in the entire image is statistically analyzed, and the standard deviation of brightness is calculated to quantify the dispersion of the image brightness values.

[0063] Set a standard deviation threshold. If the brightness standard deviation is less than the standard deviation threshold, it means that the gray values ​​of the image to be analyzed are highly concentrated and the difference between light and dark is small. It is judged as a low contrast state, that is, a gray screen. If the brightness standard deviation is greater than or equal to the standard deviation threshold, it means that the image to be analyzed has a certain level and detail and is regarded as normal contrast.

[0064] When detecting screen distortion in vehicle display systems, the high-frequency energy density of the image under analysis is analyzed by frequency domain transformation. If the high-frequency energy density is higher than the standard high-frequency energy density, it is determined to be screen distortion.

[0065] Screen flickering refers to abnormal display phenomena caused by abnormal video signal transmission, such as a loose DP interface, lost data packets, or a faulty graphics card driver, resulting in messy pixels, striped interference, mosaic, color misalignment, or random noise on the screen.

[0066] The image to be analyzed is transformed from the spatial domain to the frequency domain using the Fast Fourier Transform, and its frequency energy distribution is analyzed. The proportion of energy in the high-frequency region to the total energy is calculated as the high-frequency energy density.

[0067] The high-frequency energy density is compared with the standard high-frequency energy density. If the high-frequency energy density is higher than the standard high-frequency energy density, it is judged as a screen distortion. If the high-frequency energy density is lower than the standard high-frequency energy density, it is judged as normal.

[0068] When performing screen freeze detection on an in-vehicle display system, the difference between multiple consecutive frames of images to be analyzed is determined, and when the difference approaches zero, it is determined to be a screen freeze.

[0069] Screen freezing refers to a phenomenon where the display screen remains frozen for an extended period and fails to refresh normally. It commonly occurs during vehicle infotainment system startup, in cases of system crashes, GPU malfunctions, operating system crashes, or unresponsive applications.

[0070] By acquiring multiple consecutive frames of images to be analyzed, the degree of difference between adjacent frames is calculated; if the difference is extremely small and continues for multiple cycles, it is determined to be a screen freeze.

[0071] When performing color cast detection on an in-vehicle display system, the average values ​​of the RGB three channels of the image to be analyzed can be calculated separately. The proportion of each channel can be calculated based on the average values ​​of the RGB three channels. If the proportion of any channel exceeds the normal range, it is determined to be a color cast.

[0072] In automotive display systems, color cast refers to the phenomenon where the screen displays abnormal color tendencies, such as a reddish, bluish, yellowish, or greenish tint. This problem is usually caused by factors such as incorrect display driver parameters, abnormal DP / HDMI signal transmission, color management module malfunction, and screen hardware aging or backlight misalignment.

[0073] By analyzing the RGB three-channel distribution of an image, the ratio difference between each channel is calculated; when the difference exceeds a reasonable range, it is determined to be a color cast.

[0074] When performing no-signal detection on an in-vehicle display system, OCR technology can be used to identify whether the image to be analyzed contains keywords indicating no signal. If so, it is determined to be in a no-signal state.

[0075] There are multiple ways to train a deep learning model to obtain a pre-trained deep learning model. In one implementation, such as... Figure 6 As shown, it includes the following steps: S601: Load the pre-trained MobileNetV2 model as the base network.

[0076] S602: Add a fully connected layer and a Dropout layer after the basic network to obtain the first basic network.

[0077] S603: Freeze the convolutional layers of the first base network to obtain the second base network.

[0078] S604: Train the second basic network to obtain the trained first model.

[0079] S605: Unfreeze the convolutional layers of the first trained model to obtain the third base network.

[0080] S606: Train the third basic network to obtain a pre-trained deep learning model.

[0081] The system captures output images using a video capture card and a DP-to-HDMI cable, generating corresponding labels. For example, it determines whether the system successfully boots to the desktop after power-on by capturing and saving the image of the desktop after power-on as a normal display. Simultaneously, it artificially creates abnormal scenarios, such as black screens, blue screens, and distorted screens, capturing images using the capture card and labeling them with corresponding anomaly tags to construct a dataset. The MobileNetV2 model is used as the base network, and fully connected layers and Dropout layers are added after the base network to obtain the first base network, adapted for the in-vehicle display system detection task.

[0082] Use ImageDataGenerator to load and preprocess the dataset to generate image data that the model can recognize. Specifically, the dataset can be processed using the following code: train_datagen = ImageDataGenerator( rescale=1. / 255, # Normalize pixel values ​​to [0,1] (improves training stability) rotation_range=28, # Randomly rotate the image (from -28° to +28°) width_shift_range=0.8, # Random horizontal shift (width ratio ≤ 80%) height_shift_range=0.2, # Random vertical translation (height percentage ≤ 20%) shear_range=0.2, # Random shear transformation (simulates perspective effect) zoom_range=0.2, # Random scaling (80%~120%) horizontal_flip=True, # Randomly flip horizontally (enhances generalization) fill_mode='nearest' # Fill new pixels: use the value of the nearest neighbor pixel ) train_generator = train_datagen.flow_from_directory( 'path_to_train_data', # Training set path (must be categorized into molecular folders) target_size=(224, 224), # Adjust the image size uniformly (such as the input size commonly used in networks like ResNet). batch_size=32, # Load 32 images per batch `class_mode='sparse'` # The label returns an integer (e.g., [0, 2, 1]). ) validation_datagen = ImageDataGenerator(rescale=1. / 255) # Normalize only, do not enhance validation_generator = validation_datagen.flow_from_directory(...) Train newly added layers, fully connected layers, and Dropout layers; frozen convolutional layers will not participate in training. #Training the model history = model.fit( train_generator, steps_per_epoch=train_generator.sanples / / train_generator.batch_size, epochs=10, validation_data=validation_generator, validation_steps=validation_generator.samples / / validation_generator.batch_size During fine-tuning, some convolutional layers of the model are unfrozen and further trained so that the model can better adapt to specific tasks: # Thawing part of the convolutional layer base_model.trainable = True #Unfreeze the last few layers of MobileNetV2 fine_tune_at=100 # From which layer should the thawing begin? for layer in base_model.layers[:fine_tune_at]: layer.trainable = False #Recompile the model model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.00001), loss='sparse_categorical_crossentropy', metrics=['accuracy']) #Continue training history_finetune = model.fit( train_generator, steps_per_epoch=train_generator.samples / / train_generator.batch_size, epochs=10, validation_data=validation_generator, validation_steps=validation_generator.samples / / validation_generator.batch_size) Please refer to Figure 7 This application embodiment also provides an application for Figure 1 The vehicle display system testing device 110 of the electronic device 100 includes: Acquisition module 111 is used to acquire images output by the vehicle display system; The processing module 112 is used to preprocess the output image to obtain the image to be analyzed; input the image to be analyzed into a pre-trained deep learning model, and output the detection result of the display status of the vehicle display system.

[0083] This application also provides an electronic device 100, which includes a processor 130 and a memory 120. The memory 120 stores computer-executable instructions, which, when executed by the processor 130, implement the vehicle display system detection method.

[0084] This application embodiment also provides a computer-readable storage medium storing a computer program, which, when executed by the processor 130, implements the vehicle display system detection method.

[0085] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0086] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part. If the function is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0087] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0088] The above descriptions are merely various embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for testing an in-vehicle display system, characterized in that, The method includes: Acquire images output from the vehicle-mounted display system; The output image is preprocessed to obtain the image to be analyzed; The image to be analyzed is input into a pre-trained deep learning model, which outputs the detection result of the display status of the vehicle display system.

2. The method according to claim 1, characterized in that, The step of preprocessing the output image to obtain the image to be analyzed includes: The output image is subjected to at least one of the following: denoising, contrast enhancement, grayscale conversion, or multi-channel feature extraction, wherein the denoising is performed using a Gaussian filtering algorithm, and the contrast enhancement is performed using a histogram equalization method.

3. The method according to claim 1, characterized in that, The step of inputting the image to be analyzed into a pre-trained deep learning model and outputting the detection result of the display status of the vehicle display system includes: Determine the average brightness of the image to be analyzed; The average brightness is compared with an adaptive threshold; When the average brightness is greater than or equal to the adaptive threshold, the display state of the vehicle display system is determined to be on. When the average brightness is less than the adaptive threshold, the display status of the vehicle display system is determined to be not normally lit.

4. The method according to claim 3, characterized in that, The method further includes: Determine the average brightness and brightness standard deviation of the image to be analyzed; Calculate the product of the brightness standard deviation and the adjustment coefficient, wherein the adjustment coefficient is used to control the sensitivity of the threshold to the brightness standard deviation; The sum of the average brightness and its product is calculated as an adaptive threshold.

5. The method according to claim 3, characterized in that, The method further includes: When the display state of the vehicle display system is not normally lit, anomaly detection is performed on the display state of the vehicle display system. The anomaly detection includes at least one of the following: gray screen detection, screen distortion detection, screen freeze detection, color cast detection, and no signal detection.

6. The method according to claim 5, characterized in that, The step of detecting anomalies in the display status of the vehicle-mounted display system includes: Determine the standard deviation of the brightness of the image to be analyzed; When the brightness standard deviation is lower than the standard deviation threshold, it is determined to be a gray screen; The high-frequency energy density of the image to be analyzed is analyzed by frequency domain transformation; When the high-frequency energy density is higher than the standard high-frequency energy density, it is determined to be a screen distortion. Determine the differences between multiple consecutive frames of the image to be analyzed; When the difference approaches zero, it is determined to be a screen freeze; Calculate the mean values ​​of the RGB three channels of the image to be analyzed; The proportion of each channel is calculated based on the average of the three RGB channels; If the ratio of any channel exceeds the normal range, it is judged as color cast; OCR technology is used to identify whether the image to be analyzed contains keywords without a signal. If so, it is determined to be a no-signal state.

7. The method according to claim 1, characterized in that, The method further includes: Load the pre-trained MobileNetV2 model as the base network; A first basic network is obtained by adding a fully connected layer and a Dropout layer after the basic network. Freeze the convolutional layers of the first base network to obtain the second base network; The second basic network is trained to obtain the first trained model; The convolutional layers of the first model after training are unfrozen to obtain the third base network; The third basic network is trained to obtain a pre-trained deep learning model.

8. A vehicle-mounted display system testing device, characterized in that, The device includes: The acquisition module is used to acquire images output by the vehicle-mounted display system; The processing module is used to preprocess the output image to obtain the image to be analyzed; input the image to be analyzed into a pre-trained deep learning model, and output the detection result of the display status of the vehicle display system.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method according to any one of claims 1-7.

10. A storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method described in any one of claims 1-7.

Images