Image quality control method and device based on dual-path lens, medium and equipment

Abstract

This application discloses an image quality control method, apparatus, medium, and device based on dual-lens cameras. The method includes: setting initial focus points for registered dual lenses, and controlling the infrared lens to obtain an infrared image containing human features, and the visible light lens to obtain a visible light image containing vehicle recognition features; performing image quality evaluation on at least one human feature in the infrared image and at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result for both human and vehicle features; if both comprehensive evaluation results meet the standards, then fusing the images according to preset image fusion rules to obtain the output image from the dual lenses. This technical solution utilizes dual lenses to focus on human and vehicle recognition features separately, and fuses images that meet the quality standards, achieving optimal results for both human and vehicle recognition features in mixed-traffic scenarios, thus improving the image quality of mixed-traffic images.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to an image quality control method, apparatus, medium and device based on dual-lens cameras. Background Technology

[0002] With the rapid development of image processing technology, people are paying more and more attention to image quality, and therefore have put forward higher requirements for image quality.

[0003] Especially in nighttime environments, poor lighting poses a significant challenge to image quality. Traditional methods often utilize visible light paths to capture images of mixed pedestrian and vehicle traffic at night, but the resulting images are of poor quality. Therefore, improving the quality of mixed-traffic images is an urgent problem to be solved.

[0004] Because visible light cannot simultaneously identify human and vehicle features in a nighttime environment under a given lighting condition, it cannot achieve the same image quality for both human and vehicle features. If the image is too dark, human features will be too dark, resulting in an unclear image; if the image is too bright, vehicle features cannot be identified because the headlights cannot be suppressed. Summary of the Invention

[0005] This application provides an image quality control method, apparatus, medium, and device based on dual-lens cameras. It can use dual lenses to obtain infrared images containing human features and visible light images containing vehicle identification features, and then fuse the two images with acceptable human and vehicle identification feature quality to improve the quality of mixed-image processing.

[0006] In a first aspect, embodiments of this application provide an image quality control method based on a dual-lens system, the method comprising:

[0007] Preliminary focusing points are set for the registered dual-channel lenses, and the infrared lens in the dual-channel lenses is controlled to focus on human features to obtain an infrared image containing human features, and the visible light lens focuses on vehicle identification features to obtain a visible light image containing vehicle identification features.

[0008] Image quality evaluation is performed on at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature; and image quality evaluation is performed on at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition feature.

[0009] If the comprehensive evaluation result of the human body features meets the standard, and the comprehensive evaluation result of the vehicle recognition features meets the standard, then the images are fused according to the preset image fusion rules to obtain the output image of the dual-lens camera.

[0010] Secondly, embodiments of this application provide an image quality control device based on a dual-lens system, the device comprising:

[0011] The focus control module is used to set the initial focus points for the registered dual-channel lenses respectively, and control the infrared lens in the dual-channel lenses to focus on human features to obtain an infrared image containing human features, and the visible light lens to focus on vehicle recognition features to obtain a visible light image containing vehicle recognition features.

[0012] The image quality evaluation module is used to evaluate the image quality of at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature; and to evaluate the image quality of at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition feature.

[0013] The image fusion module is used to fuse images according to a preset image fusion rule if the comprehensive evaluation results of the human body features and the comprehensive evaluation results of the vehicle recognition features meet the standards, thereby obtaining the output images of the dual-lens cameras.

[0014] Thirdly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the image quality control method based on dual-lens as described in embodiments of this application.

[0015] Fourthly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the image quality control method based on dual-lens as described in embodiments of this application.

[0016] The technical solution provided in this application sets initial focus points for registered dual-channel lenses, and controls the infrared lens in the dual-channel lenses to focus on human features to obtain an infrared image containing human features, and the visible light lens to focus on vehicle recognition features to obtain a visible light image containing vehicle recognition features. Then, image quality evaluation is performed on at least one human feature in the infrared image to obtain a comprehensive evaluation result for human features, and image quality evaluation is performed on at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result for vehicle recognition features. If both the comprehensive evaluation results for human features and vehicle recognition features meet the standards, they are fused according to preset image fusion rules to obtain the output image of the dual-channel lenses. Using this technical solution, dual-channel lenses can be used to obtain infrared images containing human features and visible light images containing vehicle recognition features, and the two images after the quality of human features and vehicle recognition features meets the standards can be fused. This achieves a balance between the best effects of human features and vehicle recognition features in mixed-traffic scenarios, and allows for timely adjustments based on the image quality evaluation results of human features and vehicle recognition features, enabling mixed-traffic images to achieve the best effect in the current environment, thereby improving the quality of mixed-traffic images. Attached Figure Description

[0017] Figure 1 This is a flowchart of the image quality control method based on dual-lens provided in Embodiment 1 of this application;

[0018] Figure 2 This is a flowchart of the image quality control method based on dual-lens provided in Embodiment 2 of this application;

[0019] Figure 3 This is an image showing how the brightness weight changes with the brightness of human body features, as provided in Embodiment 2 of this application;

[0020] Figure 4 This is a flowchart of the image quality control method based on dual-lens provided in Embodiment 3 of this application;

[0021] Figure 5 This is an image showing how the brightness weight changes with the brightness of the vehicle recognition feature, as provided in Embodiment 3 of this application;

[0022] Figure 6 This is a flowchart of a preferred image quality control method based on dual-lens cameras provided in Embodiment 3 of this application;

[0023] Figure 7 This is a structural block diagram of an image quality control device based on a dual-lens camera provided in Embodiment 4 of the present invention;

[0024] Figure 8 This is a schematic diagram of the structure of an electronic device provided in Embodiment Six of this application. Detailed Implementation

[0025] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present application, not the entire structure.

[0026] Before discussing the exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processes, many of these steps can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the steps can be rearranged. The process can be terminated when its operation is complete, but may also have additional steps not included in the figures. The process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0027] Example 1

[0028] Figure 1 This is a flowchart of an image quality control method based on dual-lens cameras provided in Embodiment 1 of this application. This embodiment is applicable to scenarios that need to balance the best effects of human features and vehicle recognition features in mixed traffic scenarios. This method can be executed by the image quality control device based on dual-lens cameras provided in this application embodiment. The device can be implemented by software and / or hardware and can be integrated into electronic devices.

[0029] like Figure 1 As shown, the image quality control method based on dual-lens cameras includes:

[0030] S110 sets preliminary focus points for the registered dual-channel lenses respectively, and controls the infrared lens in the dual-channel lenses to focus on human features to obtain an infrared image containing human features, and the visible light lens to focus on vehicle recognition features to obtain a visible light image containing vehicle recognition features.

[0031] Dual-path lenses refer to zoom lenses that include both infrared and visible light lenses. Specifically, the infrared lens uses an infrared lamp to capture infrared images containing human features and outputs them as black and white images. The visible light lens uses a white light lamp to capture visible light images containing vehicle identification features and outputs them as color images. Human features can reflect human characteristics, such as a face. Vehicle identification features can reflect vehicle characteristics, such as a license plate. Furthermore, infrared images can contain one or more human feature regions, and similarly, visible light images can contain one or more vehicle identification feature regions.

[0032] In this embodiment, dual-lens cameras can focus on human features and vehicle recognition features separately, and simultaneously capture both features, thereby obtaining infrared and visible light images of the same time and scene. Furthermore, a hybrid illumination lamp can be used to simultaneously provide infrared and white light illumination, and a gimbal can be connected to allow for 360-degree rotation of the dual-lens cameras to adjust their viewing angle.

[0033] In this solution, after installation, the dual-lens cameras can be registered to ensure consistent field of view, preparing for subsequent dual-lens image fusion. The optimal focal points of the dual lenses can be calculated and set for initial focusing. Specifically, focusing information can be sent separately to the infrared and visible light lenses, with the infrared lens focusing on human features and the visible light lens focusing on vehicle recognition features. The optimal focal points of the two lenses can be the same or different.

[0034] In this solution, when multiple targets appear in a scene, depth information of multiple targets can be obtained through binocular parallax for assisted focusing, but it is not limited to this focusing method. Furthermore, if the initial focusing result is not optimal, i.e., the clarity of human features or vehicle recognition features does not meet the requirements, refocusing can be achieved through infrared or visible light paths to simultaneously and clearly focus multiple human features or multiple vehicle recognition features.

[0035] S120, perform image quality evaluation on at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature; and perform image quality evaluation on at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition feature.

[0036] Image quality evaluation can be used to assess the quality of human feature regions in infrared images or vehicle recognition feature regions in visible light images. These human feature regions or vehicle recognition feature regions can be rectangular areas cropped after human or vehicle features are identified in the image, and there can be one or more such regions. For example, a formula can be used to calculate the image quality of each human feature region or vehicle recognition feature region to obtain an evaluation value. These evaluation values ​​are then weighted and summed to obtain a comprehensive evaluation value for the human feature regions or vehicle recognition feature regions. This comprehensive evaluation value is then used as the comprehensive evaluation result for either the human feature region or the vehicle recognition feature region.

[0037] In this solution, evaluation criteria can be set to comprehensively evaluate the image quality of infrared and visible light images based on the comprehensive evaluation results of human features or vehicle recognition features, respectively.

[0038] In this scheme, the evaluation criteria for infrared images can be set as follows:

[0039]

[0040] Where, τ 人 This represents the threshold for meeting the standards for human body characteristic quality. In other words, when the overall human body characteristic evaluation result Ω is lower than the threshold τ... 人 When the infrared image quality is below standard, it can be considered that the overall human feature evaluation result Ω is higher than the threshold τ. 人 At this point, the infrared image quality can be considered to meet the standards.

[0041] In this scheme, the evaluation criteria for visible light images can be set as follows:

[0042]

[0043] Where, τ 车 This represents the threshold at which vehicle recognition feature quality meets the standard. In other words, when the overall evaluation result of vehicle recognition features, T, is lower than the threshold τ, the vehicle recognition feature quality is considered acceptable. 车 When the visible light image quality is below standard, it can be considered that the image quality is substandard. Conversely, when the comprehensive evaluation result of vehicle recognition features, T, exceeds the threshold τ... 车 At this point, the visible light image quality can be considered to meet the standard.

[0044] Furthermore, when the quality of infrared or visible light images is substandard, the image quality can be adjusted by changing parameters such as exposure value and fill light intensity to achieve the best results for infrared or visible light images.

[0045] S130, if the comprehensive evaluation results of human features and the comprehensive evaluation results of vehicle recognition features meet the standards, then the images are fused according to the preset image fusion rules to obtain the output images of the dual-lens cameras.

[0046] The preset image fusion rule refers to the pre-defined criteria for fusing infrared and visible light images. This can be achieved by dividing the acquired image into several image regions and performing targeted image fusion according to the preset rule. It is understood that this solution fuses infrared and visible light images that meet quality standards. Regions other than those with human or vehicle recognition features can be fused using a different method. For example, if the overall effect of the infrared path in a certain region is slightly better than the visible light path, then the proportion of the infrared image is higher than that of the visible light image. The specific proportion can be calculated by comparing parameters such as clarity, brightness, and frequency band of the two images, ultimately using a weighted sum.

[0047] The technical solution provided in this application sets initial focus points for the registered dual-lens system and controls the infrared lens in the dual-lens system to focus on human features to obtain an infrared image containing human features, and the visible light lens to focus on vehicle recognition features to obtain a visible light image containing vehicle recognition features. Then, image quality is evaluated for at least one human feature in the infrared image to obtain a comprehensive evaluation result for human features, and image quality is evaluated for at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result for vehicle recognition features. If both the comprehensive evaluation results for human features and vehicle recognition features meet the standards, they are fused according to preset image fusion rules to obtain the output image from the dual-lens system. This solution, through the above methods, can achieve optimal results for both human features and vehicle recognition features in mixed-traffic scenarios, and can make timely adjustments based on the quality evaluation results of human features and vehicle recognition features to achieve the best effect in the current environment, thereby improving the quality of mixed-traffic images.

[0048] In this embodiment, optionally, the images are fused according to a preset image fusion rule to obtain the output images of the dual-lens cameras, including:

[0049] If the region is a human body feature area, then the infrared image is fused with a fusion ratio of 100%.

[0050] If the region is a vehicle recognition feature area, then the fusion is performed with a fusion ratio of 100% for the visible light image;

[0051] If the region is neither a human body feature region nor a vehicle recognition feature region, it will be calculated as follows:

[0052]

[0053] in, The weights are respectively for brightness, sharpness, and frequency band.

[0054] Where Ψ is the fusion ratio of the infrared images, and αcolor α represents the brightness of a visible light image. IR γ represents the brightness of the infrared image. color For the contrast of a visible light image, γ IR ε represents the contrast of an infrared image. color For the sharpness of visible light images, ε IR The clarity of the infrared image is considered. The evaluation dimensions for the fusion ratio are not limited to those mentioned above.

[0055] In this solution, the fusion ratio refers to the proportion of human or vehicle recognition features in the fused image during image fusion. By overlaying and fusing dual-channel images, a real-time output of the fused live image can be generated. This solution allows for setting image fusion rules for different types of image regions, effectively improving the quality of the fused image and enhancing the applicability of image fusion.

[0056] Example 2

[0057] Figure 2 This is a flowchart of an image quality control method based on a dual-lens camera provided in Embodiment 2 of this application. This embodiment is an optimization based on the above embodiment. Specifically, the optimization involves: evaluating the image quality of at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature, including: performing human feature recognition on the infrared image to obtain at least one human feature region in the infrared image; for each human feature region, acquiring at least one of brightness, contrast, sharpness, and noise; determining the evaluation result of each human feature region based on at least one of brightness, contrast, sharpness, and noise; and performing image quality evaluation based on the evaluation results of each human feature region to obtain a comprehensive evaluation result of the human feature.

[0058] like Figure 2 As shown, the method in this embodiment specifically includes the following steps:

[0059] S210 sets preliminary focus points for the registered dual-channel lenses respectively, and controls the infrared lens in the dual-channel lenses to focus on human features to obtain an infrared image containing human features, and the visible light lens to focus on vehicle recognition features to obtain a visible light image containing vehicle recognition features.

[0060] S220, perform human feature recognition on the infrared image to obtain at least one human feature region in the infrared image.

[0061] In this solution, human features can be identified from infrared images using a human feature recognition algorithm. For example, the human feature recognition algorithm can be the eigenface method or an image recognition method based on neural networks.

[0062] S230 acquires at least one of brightness, contrast, sharpness, and noise for each human body feature area.

[0063] Among them, brightness, contrast, sharpness, and noise can be used to evaluate the image quality of human feature regions.

[0064] In this embodiment, optionally, the brightness, contrast, sharpness, and noise of the human body feature area can be calculated using the following formula:

[0065]

[0066] Where α(f) is the brightness of the human body feature region, X is the number of pixel rows in the human body feature region, Y is the number of pixel columns in the human body feature region, and f(x,y) is the brightness value of a pixel within the human body feature region. The average brightness of the human body's characteristic regions;

[0067]

[0068] Wherein, β(f) is the contrast of the human body feature region;

[0069] γ(f)=∑ y ∑ x (|f(x+1,y)-f(x,y)| 2 +|f(x,y+1)-f(x,y)| 2 );

[0070] Wherein, γ(f) represents the sharpness of the human body feature region;

[0071]

[0072] Where δ(f) represents the noise in the human body feature region.

[0073] This solution allows for the direct calculation of brightness, contrast, sharpness, and noise in human feature regions using formulas, thereby obtaining image quality evaluation parameters for human feature regions in a simple and convenient manner.

[0074] S240 determines the evaluation results for each human body feature region based on at least one of brightness, contrast, sharpness, and noise.

[0075] In this scheme, each human feature region in an infrared image can be evaluated according to a pre-set human feature region evaluation standard based on one or more parameters, including brightness, contrast, sharpness, and noise, thereby obtaining the evaluation result for each human feature region. Specifically, a standard range can be set for brightness, contrast, sharpness, and noise, and the image quality can be evaluated based on the set standard range. It is understandable that if the image quality of the human feature region is within the standard range, it meets the standard and receives a higher evaluation; conversely, if the image quality of the human feature region is below or above the standard range, it does not meet the standard and receives a lower evaluation.

[0076] S250 evaluates image quality based on the evaluation results of each human feature region, and obtains a comprehensive evaluation result of human features.

[0077] In this scheme, the evaluation results of each human feature region can be integrated to achieve the quality evaluation of infrared images and obtain a comprehensive evaluation result of human features.

[0078] In this embodiment, optionally, the comprehensive evaluation result of human features can be calculated using the following formula:

[0079]

[0080] Among them, Ω M This represents the comprehensive evaluation result of human body characteristics, where M is the number of human body characteristics. Let be the brightness of the i-th human feature region. Let the contrast of the i-th human feature region be denoted as . For the clarity of the i-th human feature region, Let μ1 be the noise of the i-th human feature region; μ2, μ3 and μ4 are the brightness weight, contrast weight, sharpness weight and noise weight of the human feature region, respectively.

[0081] In this embodiment, a comprehensive evaluation result of human features can be obtained by weighted summation of the evaluation results of each human feature region in the infrared image.

[0082] In this embodiment, optionally, if the brightness of the human feature area is not within the preset brightness range, the brightness weight decreases as the deviation between the brightness of the human feature area and the preset brightness range increases; if the contrast of the human feature area is not within the preset contrast range, the contrast weight decreases as the deviation between the contrast of the human feature area and the preset contrast range increases; if the sharpness of the human feature area is not within the preset sharpness range, the sharpness weight decreases as the deviation between the sharpness of the human feature area and the preset sharpness range increases; and the noise weight decreases as the noise in the human feature area increases.

[0083] Among them, the preset brightness range, preset contrast range, and preset sharpness range can refer to the standard brightness range, standard contrast range, and standard sharpness range preset for human body feature areas, respectively, and can be used to characterize the range of brightness weight maintenance μ1, contrast weight maintenance μ2, and sharpness weight maintenance μ3 for human body feature areas, respectively.

[0084] Understandably, the brightness weight, contrast weight, sharpness weight, and noise weight exhibit certain patterns of change. Specifically, taking the brightness of the human body feature area as an example, when the brightness of the human body feature area is within the preset brightness range, the brightness weight remains constant at μ1; when the brightness of the human body feature area is less than the lower limit of the preset brightness range, the brightness weight increases from 0 to μ1 as the brightness increases; when the brightness of the human body feature area is higher than the upper limit of the preset brightness range, the brightness weight decreases from μ1 to 0 as the brightness increases. Furthermore, the variation patterns of the contrast weight μ2 and sharpness weight μ3 of the human body feature area are basically consistent with the variation pattern of the brightness weight μ1, while the noise weight μ4 decreases as the noise in the human body feature area increases.

[0085] Figure 3 This is an image showing the change in brightness weight as a function of human body feature brightness, as provided in Embodiment 2 of this application. Here, `min` represents the minimum value of the preset brightness range for the human body feature region, and `max` represents the maximum value of the preset brightness range for the human body feature region. Figure 3 As shown, during the process of gradually increasing the brightness of human features from 0, the brightness weight successively exhibits the characteristics of gradually increasing from 0 to μ1, keeping μ1 unchanged, and gradually decreasing from μ1 to 0.

[0086] This scheme, through such a setting, integrates the evaluation results of each human feature region by adjusting the weight values, and uses the integrated result as the comprehensive evaluation result of human features, thereby enabling flexible multi-dimensional comprehensive evaluation of infrared image quality.

[0087] S260, perform image quality evaluation on at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition feature.

[0088] S270 If the comprehensive evaluation results of human features and vehicle recognition features meet the standards, then the images are fused according to the preset image fusion rules to obtain the output images of the dual-lens cameras.

[0089] The technical solution provided in this application embodiment can directly calculate the brightness, contrast, clarity, and noise of human feature areas through formulas, and then obtain a comprehensive evaluation result of human features based on at least one parameter among the brightness, contrast, clarity, and noise of human feature areas. This enables flexible multi-dimensional comprehensive evaluation of infrared image quality and further improves infrared image quality.

[0090] Example 3

[0091] Figure 4 This is a flowchart of an image quality control method based on a dual-lens camera provided in Embodiment 3 of this application. This embodiment is an optimization based on Embodiment 1 described above. Specifically, the optimization involves: performing image quality evaluation on at least one vehicle identification feature in the visible light image to obtain a comprehensive evaluation result for the vehicle identification feature, including: identifying vehicle identification features in the visible light image to obtain at least one vehicle identification feature region in the visible light image; for each vehicle identification feature region, obtaining at least one of brightness, contrast, and sharpness; determining the evaluation result of each vehicle identification feature region based on at least one of brightness, contrast, and sharpness; and performing image quality evaluation based on the evaluation results of each vehicle identification feature region to obtain a comprehensive evaluation result for the vehicle identification feature.

[0092] like Figure 4 As shown, the method in this embodiment specifically includes the following steps:

[0093] S410 sets preliminary focus points for the registered dual-channel lenses and controls the infrared lens in the dual-channel lenses to focus on human features to obtain an infrared image containing human features, and the visible light lens to focus on vehicle recognition features to obtain a visible light image containing vehicle recognition features.

[0094] S420 evaluates the image quality of at least one human feature in the infrared image and obtains a comprehensive evaluation result of the human feature.

[0095] S430, perform vehicle identification feature recognition on the visible light image to obtain at least one vehicle identification feature region in the visible light image.

[0096] In this solution, vehicle identification features can be recognized from visible light images using image recognition algorithms. For example, the image recognition algorithm can be a nonlinear dimensionality reduction image recognition method or a neural network-based image recognition method.

[0097] S440 identifies at least one of brightness, contrast, and sharpness for each vehicle's characteristic region.

[0098] Brightness, contrast, and sharpness can be used to evaluate the image quality of vehicle recognition feature areas.

[0099] In this embodiment, optionally, the brightness, contrast, and sharpness of the vehicle recognition feature area can be calculated using the following formula:

[0100]

[0101] Where α(f) is the brightness of the vehicle recognition feature region, X is the number of pixel rows in the vehicle recognition feature region, Y is the number of pixel columns in the vehicle recognition feature region, and f(x,y) is the brightness value of a pixel within the vehicle recognition feature region. The average brightness value of the vehicle recognition feature region;

[0102]

[0103] Where β(f) is the contrast of the vehicle recognition feature region;

[0104]

[0105] Where ε(f) represents the clarity of the vehicle recognition feature region, and i represents one of the eight neighborhood directions.

[0106] In this scheme, when calculating the sharpness of the vehicle recognition feature region, greater emphasis is placed on the sharpness of the image edges. The pixel neighborhood gradient method can be used to calculate the gradient in the selected neighborhood direction. Here, the neighborhood direction can refer to the horizontal, vertical, 45-degree, and 135-degree directions.

[0107] This solution allows for the direct calculation of the brightness, contrast, and sharpness of the vehicle recognition feature area using formulas, thereby obtaining image quality evaluation parameters for the vehicle recognition feature area. This method is simple and convenient.

[0108] S450 determines the evaluation result of each vehicle recognition feature region based on at least one of brightness, contrast, and sharpness.

[0109] In this scheme, each vehicle recognition feature region in a visible light image can be evaluated based on one or more parameters, including brightness, contrast, and sharpness, according to a pre-set evaluation standard. This yields an evaluation result for each vehicle recognition feature region. Specifically, a standard range can be set for brightness, contrast, and sharpness, and the image quality can be evaluated based on this standard range. Understandably, if the image quality of a vehicle recognition feature region falls within the standard range, it meets the standard and receives a higher evaluation; conversely, if the image quality is below or above the standard range, it does not meet the standard and receives a lower evaluation.

[0110] S460 evaluates image quality based on the evaluation results of each vehicle recognition feature region, and obtains a comprehensive evaluation result of vehicle recognition features.

[0111] In this scheme, the evaluation results of each vehicle recognition feature region can be integrated to achieve the quality evaluation of visible light images and obtain a comprehensive evaluation result of vehicle recognition features.

[0112] In this embodiment, optionally, the comprehensive evaluation result of vehicle recognition features can be calculated using the following formula:

[0113]

[0114] Among them, T N The result represents the comprehensive evaluation of vehicle recognition features, where N is the number of vehicle recognition features. Let be the brightness of the i-th vehicle recognition feature region. Let the contrast of the i-th vehicle recognition feature region be . Let λ be the clarity of the i-th vehicle recognition feature region; λ1, λ2, and λ3 are the brightness weight, contrast weight, and clarity weight of the vehicle recognition feature region, respectively.

[0115] In this embodiment, optionally, if the brightness of the vehicle recognition feature area is not within a preset brightness range, the brightness weight decreases as the deviation between the brightness of the vehicle recognition feature area and the preset brightness range increases; if the contrast of the vehicle recognition feature area is not within a preset contrast range, the contrast weight decreases as the deviation between the contrast of the vehicle recognition feature area and the preset contrast range increases; if the clarity of the vehicle recognition feature area is not within a preset clarity range, the clarity weight decreases as the deviation between the clarity of the vehicle recognition feature area and the preset clarity range increases.

[0116] Among them, the preset brightness range, preset contrast range, and preset sharpness range can refer to the standard brightness range, standard contrast range, and standard sharpness range preset for the vehicle recognition feature area, respectively. They can be used to characterize the range of brightness weight maintenance λ1, contrast weight maintenance λ2, and sharpness weight maintenance λ3 of the vehicle recognition feature area.

[0117] Understandably, the brightness weight, contrast weight, and sharpness weight exhibit certain patterns of change. Specifically, taking the brightness of the vehicle recognition feature area as an example, when the brightness of the vehicle recognition feature is within a preset brightness range, the brightness weight remains constant at λ1; when the brightness of the vehicle recognition feature is less than the lower limit of the preset brightness range, the brightness weight increases from 0 to λ1 as the brightness increases; when the brightness of the vehicle recognition feature is higher than the upper limit of the preset brightness range, the brightness weight decreases from λ1 to 0 as the brightness increases. Furthermore, the changing patterns of the contrast weight λ2 and sharpness weight λ3 of the vehicle recognition feature area are basically consistent with the changing pattern of the brightness weight λ1.

[0118] Figure 5 This is an image showing the change in brightness weight as a function of the vehicle recognition feature brightness, as provided in Embodiment 3 of this application. Here, Min represents the minimum value of the preset brightness range for the vehicle recognition feature region, and Max represents the maximum value of the preset brightness range for the vehicle recognition feature region. Figure 5 As shown, during the process of gradually increasing the brightness of the vehicle recognition feature from 0, the brightness weight successively exhibits the characteristics of gradually increasing from 0 to λ1, keeping λ1 unchanged, and gradually decreasing from λ1 to 0.

[0119] This scheme, through such a setting, integrates the evaluation results of each vehicle recognition feature region by adjusting the weight values, and uses the integrated result as the comprehensive evaluation result of vehicle recognition features, thereby enabling flexible multi-dimensional comprehensive evaluation of visible light image quality.

[0120] S470 If the comprehensive evaluation results of human features and vehicle recognition features meet the standards, then the images are fused according to the preset image fusion rules to obtain the output images of the dual-lens cameras.

[0121] The technical solution provided in this application embodiment can directly calculate the brightness, contrast, and sharpness of the vehicle recognition feature area through formulas, and then obtain a comprehensive evaluation result of the vehicle recognition feature based on at least one of the parameters of brightness, contrast, and sharpness of the vehicle recognition feature area. This enables flexible multi-dimensional comprehensive evaluation of the visible light image quality and further improves the visible light image quality.

[0122] Figure 6 This is a flowchart of a preferred image quality control method based on dual-lens cameras, provided in Embodiment 3 of this application. Specifically, human features are human faces, vehicle recognition features are license plates, and the image optimization rule refers to the optimal fusion rule selected from preset image fusion rules based on the characteristics of the image region. When the image quality is deemed substandard, the infrared or visible light image can be optimized by adjusting peripherals and image parameters.

[0123] This solution, through such settings, can adjust peripherals and image parameters to achieve the best effect for substandard images, and can select the optimal fusion rule based on the characteristics of the image region, thereby further improving the quality of mixed images.

[0124] Example 4

[0125] Figure 7 This is a structural block diagram of an image quality control device based on a dual-lens camera, provided in Embodiment 4 of the present invention. This device can execute the image quality control method based on a dual-lens camera provided in any embodiment of the present invention, and possesses the corresponding functional modules and beneficial effects for executing the method. For example... Figure 7 As shown, the device may include:

[0126] The focusing control module 710 is used to set the initial focusing points of the registered dual-channel lenses respectively, and control the infrared lens in the dual-channel lenses to focus on human features to obtain an infrared image containing human features, and the visible light lens to focus on vehicle recognition features to obtain a visible light image containing vehicle recognition features.

[0127] The image quality evaluation module 720 is used to evaluate the image quality of at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature; and to evaluate the image quality of at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition feature.

[0128] The image fusion module 730 is used to fuse images according to a preset image fusion rule if the comprehensive evaluation result of the human body features meets the standard and the comprehensive evaluation result of the vehicle recognition features meets the standard, so as to obtain the output image of the dual-lens camera.

[0129] Based on the above embodiments, optionally, the image quality evaluation module 720 is specifically used for:

[0130] Human feature recognition is performed on the infrared image to obtain at least one human feature region in the infrared image;

[0131] For each human body feature area, acquire at least one of the following: brightness, contrast, sharpness, and noise.

[0132] The evaluation results for each human body feature region are determined based on at least one of the brightness, contrast, sharpness, and noise.

[0133] Image quality is evaluated based on the evaluation results of each human feature region to obtain a comprehensive evaluation result of human features.

[0134] Based on the above embodiments, optionally, the comprehensive evaluation result of human characteristics can be calculated using the following formula:

[0135]

[0136] Among them, Ω M This represents the comprehensive evaluation result of human body characteristics, where M is the number of human body characteristics. Let be the brightness of the i-th human feature region. Let the contrast of the i-th human feature region be denoted as . For the clarity of the i-th human feature region, Let μ1 be the noise of the i-th human feature region; μ2, μ3 and μ4 are the brightness weight, contrast weight, sharpness weight and noise weight of the human feature region, respectively.

[0137]

[0138] Where α(f) is the brightness of the human body feature region, X is the number of pixel rows in the human body feature region, Y is the number of pixel columns in the human body feature region, and f(x,y) is the brightness value of a pixel within the human body feature region. The average brightness of the human body's characteristic regions;

[0139]

[0140] Wherein, β(f) is the contrast of the human body feature region;

[0141] γ(f)=∑ y ∑ x (|f(x+1,y)-f(x,y)| 2 +|f(x,y+1)-f(x,y)| 2 );

[0142] Wherein, γ(f) represents the sharpness of the human body feature region;

[0143]

[0144] Where δ(f) represents the noise in the human body feature region.

[0145] Based on the above embodiments, optionally, if the brightness of the human body feature area is not within the preset brightness range, the brightness weight decreases as the deviation between the brightness of the human body feature area and the preset brightness range increases;

[0146] If the contrast of the human body feature region is not within the preset contrast range, the contrast weight decreases as the deviation of the contrast of the human body feature region from the preset contrast range increases.

[0147] If the clarity of the human feature region is not within the preset clarity range, the clarity weight decreases as the deviation between the clarity of the human feature region and the preset clarity range increases.

[0148] The noise weight decreases as the noise in the human body feature region increases.

[0149] Based on the above embodiments, optionally, the image quality evaluation module 720 is further used for:

[0150] Vehicle identification feature recognition is performed on the visible light image to obtain at least one vehicle identification feature region in the visible light image;

[0151] For each vehicle identification feature area, acquire at least one of brightness, contrast, and sharpness;

[0152] The evaluation result of each vehicle recognition feature region is determined based on at least one of the brightness, contrast and sharpness.

[0153] Image quality is evaluated based on the evaluation results of each vehicle recognition feature region to obtain a comprehensive evaluation result of vehicle recognition features.

[0154] Based on the above embodiments, optionally, the comprehensive evaluation result of vehicle recognition features can be calculated using the following formula:

[0155]

[0156] Among them, T N The result represents the comprehensive evaluation of vehicle recognition features, where N is the number of vehicle recognition features. Let be the brightness of the i-th vehicle recognition feature region. Let the contrast of the i-th vehicle recognition feature region be . Let λ1, λ2, and λ3 represent the brightness weight, contrast weight, and clarity weight of the vehicle recognition feature region, respectively.

[0157]

[0158] Where α(f) is the brightness of the vehicle recognition feature region, X is the number of pixel rows in the vehicle recognition feature region, Y is the number of pixel columns in the vehicle recognition feature region, and f(x,y) is the brightness value of a pixel within the vehicle recognition feature region. The average brightness value of the vehicle recognition feature region;

[0159]

[0160] Where β(f) is the contrast of the vehicle recognition feature region;

[0161]

[0162] Where ε(f) represents the clarity of the vehicle recognition feature region, and i represents one of the eight neighborhood directions.

[0163] Based on the above embodiments, optionally, if the brightness of the vehicle recognition feature area is not within the preset brightness range, the brightness weight decreases as the deviation between the brightness of the vehicle recognition feature area and the preset brightness range increases;

[0164] If the contrast of the vehicle recognition feature region is not within the preset contrast range, the contrast weight decreases as the deviation of the contrast of the vehicle recognition feature region from the preset contrast range increases.

[0165] If the clarity of the vehicle identification feature area is not within the preset clarity range, the clarity weight decreases as the deviation between the clarity of the vehicle identification feature area and the preset clarity range increases.

[0166] Based on the above embodiments, optionally, the image fusion module 730 is specifically used for:

[0167] If the region is a human body feature area, then the infrared image is fused with a fusion ratio of 100%.

[0168] If the region is a vehicle recognition feature area, then the fusion is performed with a fusion ratio of 100% for the visible light image;

[0169] If the region is neither a human body feature region nor a vehicle recognition feature region, it will be calculated as follows:

[0170]

[0171] in, The weights are respectively for brightness, sharpness, and frequency band.

[0172] Where Ψ is the fusion ratio of the infrared images, and α color α represents the brightness of a visible light image. IR γ represents the brightness of the infrared image. color For the contrast of a visible light image, γ IR ε represents the contrast of an infrared image. color For the sharpness of visible light images, ε IR For the clarity of infrared images.

[0173] The above-mentioned product can perform the image quality control method based on dual-lens provided in the embodiments of this application, and has the corresponding functional modules and beneficial effects of the method.

[0174] Example 5

[0175] Embodiment 5 of the present invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the image quality control method based on a dual-lens camera as provided in all embodiments of the present application:

[0176] Preliminary focusing points are set for the registered dual-channel lenses, and the infrared lens in the dual-channel lenses is controlled to focus on human features to obtain an infrared image containing human features, and the visible light lens focuses on vehicle identification features to obtain a visible light image containing vehicle identification features.

[0177] Image quality evaluation is performed on at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature; and image quality evaluation is performed on at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition feature.

[0178] If the comprehensive evaluation result of the human body features meets the standard, and the comprehensive evaluation result of the vehicle recognition features meets the standard, then the images are fused according to the preset image fusion rules to obtain the output image of the dual-lens camera.

[0179] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0180] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including—but not limited to—electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0181] The program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0182] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0183] Example 6

[0184] Embodiment 6 of this application provides an electronic device. Figure 8 This is a schematic diagram of the structure of an electronic device provided in Embodiment Six of this application. Figure 8 As shown, this embodiment provides an electronic device 800, which includes: one or more processors 820; and a storage device 810 for storing one or more programs. When the one or more programs are executed by the one or more processors 820, the one or more processors 820 implement the image quality control method based on dual-lens provided in this application embodiment. The method includes:

[0185] Preliminary focusing points are set for the registered dual-channel lenses, and the infrared lens in the dual-channel lenses is controlled to focus on human features to obtain an infrared image containing human features, and the visible light lens focuses on vehicle identification features to obtain a visible light image containing vehicle identification features.

[0186] Image quality evaluation is performed on at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature; and image quality evaluation is performed on at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition feature.

[0187] If the comprehensive evaluation result of the human body features meets the standard, and the comprehensive evaluation result of the vehicle recognition features meets the standard, then the images are fused according to the preset image fusion rules to obtain the output image of the dual-lens camera.

[0188] Of course, those skilled in the art will understand that the processor 820 also implements the technical solution of the image quality control method based on dual-lens provided in any embodiment of this application.

[0189] Figure 8 The electronic device 800 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0190] like Figure 8 As shown, the electronic device 800 includes a processor 820, a storage device 810, an input device 830, and an output device 840; the number of processors 820 in the electronic device can be one or more. Figure 8 Taking a processor 820 as an example; the processor 820, storage device 810, input device 830, and output device 840 in an electronic device can be connected via a bus or other means. Figure 8 Taking the connection between China and Israel via bus 850 as an example.

[0191] The storage device 810, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and module units, such as the program instructions corresponding to the image quality control method based on dual-lens in the embodiments of this application.

[0192] Storage device 810 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, storage device 810 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, storage device 810 may further include memory remotely located relative to processor 820, and this remote memory may be connected via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0193] Input device 830 can be used to receive input digital, character, or voice information, and to generate key signal inputs related to user settings and function control of the electronic device. Output device 840 may include electronic devices such as a display screen and a speaker.

[0194] The image quality control device, medium, and electronic device based on a dual-lens system provided in the above embodiments can execute the image quality control method based on a dual-lens system provided in any embodiment of this application, and have the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in the above embodiments can be found in the image quality control method based on a dual-lens system provided in any embodiment of this application.

[0195] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. An image quality control method based on dual-lens cameras, characterized in that, The method includes: Preliminary focusing points are set for the registered dual-channel lenses, and the infrared lens in the dual-channel lenses is controlled to focus on human features to obtain an infrared image containing human features, and the visible light lens focuses on vehicle identification features to obtain a visible light image containing vehicle identification features. Image quality evaluation is performed on at least one human feature in the infrared image to obtain a comprehensive evaluation result of human features; and image quality evaluation is performed on at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of vehicle recognition features. The comprehensive evaluation result of human features is obtained by calculating the image quality of human feature regions using a set formula, obtaining the evaluation value of each human feature region, and then weighting and summing the evaluation values ​​to obtain a comprehensive evaluation value of the human feature region. The human feature region is a rectangular area cropped after human features are recognized in the image. If the comprehensive evaluation result of the human body features meets the standard, and the comprehensive evaluation result of the vehicle recognition features meets the standard, then the images are fused according to the preset image fusion rules to obtain the output image of the dual-lens camera. The process of fusing images according to preset image fusion rules includes: The obtained infrared image and the visible light image are divided into several image regions, and different types of image regions are fused according to the preset image fusion rules. The preset image fusion rules are used to indicate the fusion ratio of the infrared image and the visible light image during the image fusion process, and the fusion ratio is determined based on the regional feature type to which the image region belongs. The regional feature types include: human feature regions, vehicle recognition feature regions, and non-human feature regions and non-vehicle recognition feature regions.

2. The method according to claim 1, characterized in that, Image quality evaluation is performed on at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature, including: Human feature recognition is performed on the infrared image to obtain at least one human feature region in the infrared image; For each human body feature area, acquire at least one of the following: brightness, contrast, sharpness, and noise. The evaluation results for each human body feature region are determined based on at least one of the brightness, contrast, sharpness, and noise. Image quality is evaluated based on the evaluation results of each human feature region to obtain a comprehensive evaluation result of human features.

3. The method according to claim 2, characterized in that, The comprehensive evaluation result of human characteristics is calculated using the following formula: ； in, This represents the comprehensive evaluation result of human body characteristics, where M is the number of human body characteristics. Let be the brightness of the i-th human feature region. Let the contrast of the i-th human feature region be denoted as . For the clarity of the i-th human feature region, The noise in the i-th human feature region; , , as well as These are the brightness weight, contrast weight, sharpness weight, and noise weight of the human body feature region, respectively. ； in, X represents the brightness of the human body feature region, X represents the number of pixel rows in the human body feature region, and Y represents the number of pixel columns in the human body feature region. The brightness value of a pixel within a human body feature region. The average brightness of the human body's characteristic regions; ； in, Contrast of human body feature areas; ； in, For the clarity of human body feature areas; ； in, Noise in areas characteristic of the human body.

4. The method according to claim 3, characterized in that: If the brightness of the human body feature area is not within the preset brightness range, the brightness weight decreases as the deviation between the brightness of the human body feature area and the preset brightness range increases. If the contrast of the human body feature region is not within the preset contrast range, the contrast weight decreases as the deviation of the contrast of the human body feature region from the preset contrast range increases. If the clarity of the human feature region is not within the preset clarity range, the clarity weight decreases as the deviation between the clarity of the human feature region and the preset clarity range increases. The noise weight decreases as the noise in the human body feature region increases.

5. The method according to claim 1, characterized in that, Image quality evaluation is performed on at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition features, including: Vehicle identification feature recognition is performed on the visible light image to obtain at least one vehicle identification feature region in the visible light image; For each vehicle identification feature area, acquire at least one of brightness, contrast, and sharpness; The evaluation result of each vehicle recognition feature region is determined based on at least one of the brightness, contrast and sharpness. Image quality is evaluated based on the evaluation results of each vehicle recognition feature region to obtain a comprehensive evaluation result of vehicle recognition features.

6. The method according to claim 5, characterized in that, The comprehensive evaluation result of vehicle recognition features is calculated using the following formula: ； in, The result represents the comprehensive evaluation of vehicle recognition features, where N is the number of vehicle recognition features. Let be the brightness of the i-th vehicle recognition feature region. Let the contrast of the i-th vehicle recognition feature region be . Let be the clarity of the i-th vehicle recognition feature region; , as well as These are the brightness weight, contrast weight, and sharpness weight of the vehicle recognition feature region, respectively. ； in, X represents the brightness of the vehicle recognition feature region, X represents the number of pixel rows in the vehicle recognition feature region, and Y represents the number of pixel columns in the vehicle recognition feature region. The brightness value of the pixel within the vehicle recognition feature area. The average brightness value of the vehicle recognition feature region; ； in, Contrast of vehicle identification feature regions; ； in, For the clarity of vehicle recognition feature regions, It represents one of the eight neighborhood directions.

7. The method according to claim 6, characterized in that, If the brightness of the vehicle recognition feature area is not within the preset brightness range, the brightness weight decreases as the deviation between the brightness of the vehicle recognition feature area and the preset brightness range increases. If the contrast of the vehicle recognition feature region is not within the preset contrast range, the contrast weight decreases as the deviation of the contrast of the vehicle recognition feature region from the preset contrast range increases. If the clarity of the vehicle identification feature area is not within the preset clarity range, the clarity weight decreases as the deviation between the clarity of the vehicle identification feature area and the preset clarity range increases.

8. The method according to claim 1, characterized in that, The images are fused according to preset image fusion rules to obtain the output images from the dual lenses, including: If the region is a human body feature area, then the infrared image is fused with a fusion ratio of 100%. If the region is a vehicle recognition feature area, then the fusion is performed with a fusion ratio of 100% for the visible light image; If the region is neither a human body feature region nor a vehicle recognition feature region, it will be calculated as follows: ； in, The weights are respectively for brightness, sharpness, and frequency band. ; in, The fusion ratio of infrared images. The brightness of the visible light image. The brightness of the infrared image. For the contrast of a visible light image, For the contrast of infrared images, For the clarity of visible light images, For the clarity of infrared images.

9. An image quality control device based on dual-lens cameras, characterized in that, The device includes: The focus control module is used to set the initial focus points for the registered dual-channel lenses respectively, and control the infrared lens in the dual-channel lenses to focus on human features to obtain an infrared image containing human features, and the visible light lens to focus on vehicle recognition features to obtain a visible light image containing vehicle recognition features. The image quality evaluation module is used to evaluate the image quality of at least one human feature in the infrared image to obtain a comprehensive evaluation result of the human feature; and to evaluate the image quality of at least one vehicle recognition feature in the visible light image to obtain a comprehensive evaluation result of the vehicle recognition feature. The comprehensive evaluation result of the human feature is obtained by calculating the image quality of the human feature region using a set formula, obtaining the evaluation value of each human feature region, and then weighting and summing the evaluation values ​​to obtain the comprehensive evaluation value of the human feature region. The human feature region is a rectangular area cropped after the human feature is recognized in the image. The image fusion module is used to fuse images according to a preset image fusion rule if the comprehensive evaluation result of the human body features meets the standard and the comprehensive evaluation result of the vehicle recognition features meets the standard, so as to obtain the output image of the dual-lens camera. The fusion according to the preset image fusion rules includes: dividing the obtained infrared image and the visible light image into several image regions, and fusing different types of image regions according to the preset image fusion rules; wherein, the preset image fusion rules are used to indicate the fusion ratio of the infrared image and the visible light image during the image fusion process, and the fusion ratio is determined based on the regional feature type to which the image region belongs, and the regional feature type includes: human feature region, vehicle recognition feature region, and non-human feature region and non-vehicle recognition feature region.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the image quality control method based on dual-lens as described in any one of claims 1-8.

11. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the image quality control method based on dual-lens as described in any one of claims 1-8.

Images