Intelligent detection system and method for lamp

Abstract

The invention relates to the technical field of lamp detection, in particular to an intelligent detection system and method for a lamp. The method comprises the following steps: acquiring brightness distribution information and an environment image information set, analyzing the brightness distribution information based on the environment image information set, and determining environment light mapping information; the environment image information set is analyzed, environment glare influence factors are analyzed and extracted, and a glare influence feature set is determined; analyzing the glare influence feature set based on the ambient light mapping information, and determining glare hidden danger distribution information; and determining and outputting a glare detection report according to the glare hidden danger distribution information. According to the method, the indirect glare risk corresponding to each area in the detection area under the current light condition is quantitatively analyzed, so that the indirect glare generated under the influence of multi-dimensional factors in the detection area can be accurately evaluated, the hidden danger of the glare can be found in time, and the negative influence of the glare on a user is remarkably reduced.

Description

Technical Field

[0001] The present application relates to the technical field of lamp detection, and in particular to an intelligent detection system and method for lamps. Background Art

[0002] In modern architectural design, comfortable lighting conditions are crucial to people's daily life and work. Glare, as a common visual obstacle, can cause visual fatigue and visual discomfort to people in a glare environment, thus having a negative impact on people's production and life.

[0003] Existing glare detection technology usually relies on the single-dimensional quantitative analysis data of light source brightness by imaging luminance meters, which makes it difficult to accurately evaluate the indirect glare caused by multi-dimensional factors. As a result, glare risks are difficult to eliminate in a timely manner, causing negative impacts on users. Summary of the invention

[0004] The present application provides an intelligent detection system and method for lamps to solve the above technical problems.

[0005] In a first aspect, the present application provides an intelligent detection method for lamps, the method comprising: obtaining brightness distribution information and an environmental image information set, analyzing the brightness distribution information based on the environmental image information set, and determining ambient light mapping information; analyzing the environmental image information set, analyzing and extracting environmental glare influencing factors, and determining a glare influence feature set; based on the ambient light mapping information, analyzing the glare influence feature set, and determining glare hazard distribution information; and determining and outputting a glare detection report based on the glare hazard distribution information.

[0006] Through this solution, the environmental image information set and the brightness distribution information are analyzed to determine the environmental light mapping information reflecting the brightness characteristics corresponding to different entity factors in the current detection area. At the same time, through the analysis of the environmental image information set, the glare influence feature set reflecting the specific influence of each environmental glare influence factor on the indirect glare is determined. Through the combined analysis of the environmental light mapping information and the glare influence feature set, the indirect glare risk corresponding to each area in the detection area under the current light conditions is quantitatively analyzed to construct the glare hidden danger distribution information. On this basis, the corresponding glare detection report is provided to the detection personnel to accurately evaluate the indirect glare generated by the multi-dimensional factors in the detection area, so that the glare hidden dangers can be discovered in time and the negative impact of glare on users can be significantly reduced.

[0007] Optionally, the step of analyzing the brightness distribution information and determining the ambient light mapping information based on the ambient image information set includes: extracting a plurality of brightness intervals based on the brightness distribution information; performing spatial coordinate mapping processing on the brightness distribution information based on the ambient image information set to determine brightness mark image information; performing brightness layering processing on the ambient image information set based on the brightness mark image information and the plurality of brightness intervals to determine a plurality of brightness layering areas; analyzing a plurality of brightness layering areas based on the brightness distribution information to determine the regional average brightness and the number of adjacent layering areas corresponding to each brightness layering area; determining the illumination transition mutation index of each brightness layering area based on the regional average brightness and the number of adjacent layering areas; and using the illumination transition mutation index of each brightness layering area as the ambient light mapping information.

[0008] Through this scheme, a number of brightness intervals are divided according to the brightness distribution information, and at the same time, the brightness distribution information is subjected to spatial coordinate mapping processing according to the environmental image information set to determine the brightness mark image information. On this basis, a number of brightness layered areas are divided, and according to the average brightness of the area corresponding to each of the brightness layered areas and the number of adjacent layered areas, the illumination transition mutation index used to reflect the degree of light transition mutation between different brightness layered areas is determined, so as to construct the ambient light mapping information, fully considering the common characteristics of light under different indirect glare conditions, so that the subsequent indirect glare analysis process based on the ambient light mapping information is more accurate and comprehensive.

[0009] Optionally, the illumination transition mutation index of each brightness layered area is determined according to the average brightness of the area and the number of adjacent layered areas, specifically as follows: ;in, For the The illumination transition mutation index of the brightness layered area, For the The average brightness of the brightness layered areas, For the The number of adjacent layered regions of the brightness layered region, For the The brightness layered area is adjacent to the The average brightness of the brightness layered areas.

[0010] Through this scheme, mathematical analysis methods are used to accurately quantify the illumination transition mutation index used to reflect whether the light transition is smooth based on the average regional brightness corresponding to each brightness stratified area and the number of adjacent stratified areas through scientific mathematical formulas, thereby improving the accuracy and scientificity of the illumination transition mutation index evaluation process, thereby improving the accuracy and analysis efficiency of the subsequent indirect glare analysis process based on the illumination transition mutation index.

[0011] Optionally, the analysis of the environmental image information set, the analysis and extraction of environmental glare influencing factors, and the determination of the glare impact feature set include: performing unified denoising and normalization processing on each environmental image in the environmental image information set according to a preset denoising and normalization strategy to determine a preprocessed image information set; based on the preprocessed image information set, performing entity recognition and edge contour extraction on objects in each preprocessed image to determine an object surface feature set and an object geometric feature information set; based on the preprocessed image information set, performing a two-stage conversion on the color space corresponding to each preprocessed image to determine the human eye adaptive color features of each object and construct an object color feature set; and constructing the glare impact feature set based on the object surface feature set, the object geometric feature information set and the color feature set.

[0012] Through this scheme, the environmental image information set is denoised and normalized to determine the preprocessed image information set. On this basis, the object surface feature set and the object geometric feature information set are determined by entity recognition and edge contour extraction of the objects in the preprocessed image. At the same time, the color space of the preprocessed image is converted in two stages so that the object color feature set is consistent with the human eye perception state. The above-mentioned object surface feature set, object geometric feature information set and color feature set are used as the glare influence feature set to comprehensively reflect the impact of environmental glare influencing factors on indirect glare.

[0013] Optionally, based on the preprocessed image information set, a two-stage conversion is performed on the color space corresponding to each preprocessed image to determine the human eye adaptive color feature of each object, specifically the following formula: ;in, is the human eye adaptive color feature corresponding to the current object pixel, is the RGB color space value corresponding to each pixel value of the current object, It is the function of converting RGB color space to XYZ color space. Converts XYZ color space to Lab color space function.

[0014] Through this solution, based on the preprocessed image information set, a two-stage conversion is performed on the color space corresponding to each preprocessed image, and the default RGB color space of the environmental image is converted into the Lab color space that conforms to the human visual experience, so as to assist in the subsequent analysis of indirect glare risks, make the glare detection results conform to the human eye perception characteristics, and improve the accuracy and scientificity of subsequent glare detection results.

[0015] Optionally, the step of analyzing the glare impact feature set based on the ambient light mapping information to determine the glare hazard distribution information includes: comparing the illumination transition mutation index of each brightness layered area in the ambient light mapping information with a preset illumination mutation threshold, extracting all corresponding brightness layered areas whose illumination transition mutation index is greater than the preset illumination mutation threshold, and determining a number of key analysis areas; screening the glare impact feature set based on the several key analysis areas, extracting the object surface feature set, the object geometric feature information set and the object surface feature set of each object in the key analysis area. The color feature set is determined; according to the object surface feature set, the surface roughness of each object is determined; the object geometric feature information set is analyzed to extract the corner sharpness of each object; according to the surface roughness, the corner sharpness and the color feature set, a surface reflection influence factor, a corner sharpness influence factor and a color gain influence factor are determined respectively; according to the brightness distribution information, an average background brightness is determined, and based on the average background brightness, a glare risk gain value corresponding to each object is determined according to the surface reflection influence factor, the corner sharpness influence factor and the color gain influence factor, so as to construct the glare hidden danger distribution information.

[0016] Through this scheme, according to the comparison result of the illumination transition mutation index and the preset illumination mutation threshold, several key analysis areas are screened, and the surface feature set, geometric feature information set and color feature set of each object in the key analysis area are extracted. On this basis, the surface roughness and edge sharpness corresponding to the object are analyzed, and then the surface reflection influencing factor, edge sharpness influencing factor and color gain influencing factor are determined. In combination with the average background brightness, the glare risk gain value of each object is evaluated, so as to construct the glare hazard distribution information, so that the analysis process of indirect glare hazards is closely combined with the characteristics of each object in the environment, improve the accuracy and pertinence of the glare hazard distribution information, and significantly improve the accuracy of indirect glare detection under the influence of multi-dimensional factors.

[0017] Optionally, based on the average background brightness, the glare risk gain value corresponding to each object is determined according to the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor, specifically as follows: ;in, is the glare risk gain value, is the surface reflection influence factor, is the edge sharpness influencing factor, is the color gain influencing factor, is the average background brightness, is the preset weighted impact index, The preset sensitivity adjustment index.

[0018] Through this scheme, mathematical analysis is used, based on the average background brightness, according to the surface reflection influence factor, edge sharpness influence factor and color gain influence factor, through a clear mathematical relationship language, to describe the influence of the above parameters on indirect glare, so as to quantify the glare risk gain value corresponding to each object, significantly improving the scientificity and accuracy of the glare risk gain value.

[0019] Optionally, the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor are determined respectively according to the surface roughness, the edge sharpness and the color feature set, specifically as the following formulas: ; in, is the surface reflection influence factor, is the preset slope factor, is the surface roughness, is the preset roughness reference value, is the edge sharpness influencing factor, is the edge sharpness, To preset the sharpness sensitivity index, is the color gain influencing factor, is the human eye adaptive color feature corresponding to the current object pixel, is the preset color gain relationship coefficient.

[0020] Through this scheme, mathematical analysis is used to scientifically quantify the surface reflection influence factor, edge sharpness influence factor and color gain influence factor according to the surface roughness, edge sharpness and color feature set, thereby significantly improving the accuracy and scientificity of the surface reflection influence factor, edge sharpness influence factor and color gain influence factor, thereby improving the accuracy of the assessment of indirect glare risk based on the above factors.

[0021] Optionally, determining and outputting a glare detection report based on the glare hazard distribution information includes: analyzing a number of the brightness stratified areas according to the glare risk gain value corresponding to each object in the glare hazard distribution information, and determining a comprehensive glare risk corresponding to each of the brightness stratified areas; marking the corresponding area in each environmental image in the environmental image information set with glare risk according to the comprehensive glare risk corresponding to each of the brightness stratified areas, and determining a glare risk distribution image set; and visually outputting the glare risk distribution image set as the glare detection report.

[0022] Through this solution, based on the glare hazard distribution information, the comprehensive glare risk corresponding to each of the brightness stratified areas is analyzed and obtained, and the corresponding area in each environmental image in the environmental image information set is marked with glare risk, so as to obtain a glare risk distribution image set that can intuitively reflect the overall indirect glare risk of each area in the environment, and the glare risk distribution image set is visualized and output as a glare detection report to improve the readability of the glare detection report.

[0023] In a second aspect, the present application provides an intelligent detection system for a lamp, the system comprising: A mapping analysis module is used to obtain brightness distribution information and an environmental image information set, analyze the brightness distribution information based on the environmental image information set, and determine the ambient light mapping information; a feature extraction module is used to analyze the environmental image information set based on the brightness distribution information, analyze and extract environmental glare influencing factors, and determine the glare influencing feature set; a hidden danger analysis module is used to analyze the glare influencing feature set based on the ambient light mapping information, and determine the glare hidden danger distribution information; an output module is used to determine and output a glare detection report based on the glare hidden danger distribution information.

[0024] Optionally, the mapping analysis module is specifically used to: extract a number of brightness intervals according to the brightness distribution information; perform spatial coordinate mapping processing on the brightness distribution information according to the environmental image information set to determine brightness mark image information; based on the brightness mark image information, perform brightness stratification processing on the environmental image information set according to the several brightness intervals to determine a number of brightness stratification areas; based on the brightness distribution information, analyze the several brightness stratification areas to determine the regional average brightness and the number of adjacent stratification areas corresponding to each brightness stratification area; determine the illumination transition mutation index of each brightness stratification area according to the regional average brightness and the number of adjacent stratification areas; and use the illumination transition mutation index of each brightness stratification area as the ambient light mapping information.

[0025] Optionally, when the mapping analysis module determines the illumination transition mutation index of each brightness layered area according to the average brightness of the area and the number of adjacent layered areas, the specific formula is as follows: ; in, For the The illumination transition mutation index of the brightness layered area, For the The average brightness of the brightness layered areas, For the The number of adjacent layered regions of the brightness layered region, For the The brightness layered area is adjacent to the The average brightness of the brightness layered areas.

[0026] Optionally, the feature extraction module is specifically used to: perform unified denoising and normalization processing on each environmental image in the environmental image information set according to a preset denoising and normalization strategy to determine a preprocessed image information set; based on the preprocessed image information set, perform entity recognition and edge contour extraction on objects in each preprocessed image to determine an object surface feature set and an object geometric feature information set; based on the preprocessed image information set, perform a two-stage conversion on the color space corresponding to each preprocessed image to determine the human eye adaptive color features of each object and construct an object color feature set; and construct the glare impact feature set based on the object surface feature set, the object geometric feature information set and the color feature set.

[0027] Optionally, when the feature extraction module performs a two-stage conversion on the color space corresponding to each preprocessed image based on the preprocessed image information set to determine the human eye adaptive color feature of each object, the specific formula is as follows: ;in, is the human eye adaptive color feature corresponding to the current object pixel, is the RGB color space value corresponding to each pixel value of the current object, It is the function of converting RGB color space to XYZ color space. Converts XYZ color space to Lab color space function.

[0028] Optionally, the hidden danger analysis module is specifically used to: compare the illumination transition mutation index of each brightness layered area in the ambient light mapping information with a preset illumination mutation threshold, extract all corresponding brightness layered areas whose illumination transition mutation index is greater than the preset illumination mutation threshold, and determine a number of key analysis areas; screen the glare impact feature set according to the several key analysis areas, extract the object surface feature set, the object geometric feature information set and the color feature set of each object in the key analysis area; determine the surface roughness of each object according to the object surface feature set; analyze the object geometric feature information set to extract the edge sharpness of each object; determine the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor according to the surface roughness, the edge sharpness and the color feature set; determine the average background brightness according to the brightness distribution information, and based on the average background brightness, determine the glare risk gain value corresponding to each object according to the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor, so as to construct the glare hidden danger distribution information.

[0029] Optionally, when the hidden danger analysis module determines the glare risk gain value corresponding to each object based on the average background brightness, according to the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor, the specific formula is as follows: ;in, is the glare risk gain value, is the surface reflection influence factor, is the edge sharpness influencing factor, is the color gain influencing factor, is the average background brightness, is the preset weighted impact index, The preset sensitivity adjustment index.

[0030] Optionally, when the hidden danger analysis module determines the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor according to the surface roughness, the edge sharpness and the color feature set, the specific formulas are as follows: ; in, is the surface reflection influence factor, is the preset slope factor, is the surface roughness, is the preset roughness reference value, is the edge sharpness influencing factor, is the edge sharpness, To preset the sharpness sensitivity index, is the color gain influencing factor, is the human eye adaptive color feature corresponding to the current object pixel, is the preset color gain relationship coefficient.

[0031] Optionally, the output module is specifically used to: analyze a number of the brightness stratified areas according to the glare risk gain value corresponding to each object in the glare hazard distribution information, and determine the comprehensive glare risk corresponding to each of the brightness stratified areas; mark the corresponding area in each environmental image in the environmental image information set with glare risk according to the comprehensive glare risk corresponding to each of the brightness stratified areas, and determine a glare risk distribution image set; and visualize and output the glare risk distribution image set as the glare detection report. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0033] Figure 1 A schematic diagram of an application scenario provided for an embodiment of the present application; Figure 2 A flow chart of an intelligent detection method for a lamp provided in one embodiment of the present application; Figure 3 A schematic diagram of the structure of an intelligent detection system for a lamp provided in one embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0035] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0036] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0037] Existing glare detection technology usually relies on the single-dimensional quantitative analysis data of light source brightness by imaging luminance meters, which makes it difficult to accurately evaluate the indirect glare caused by multi-dimensional factors. As a result, glare risks are difficult to eliminate in a timely manner, causing negative impacts on users.

[0038] Based on this, the present application provides an intelligent detection system and method for lamps. The environmental image information set and brightness distribution information are analyzed to determine the ambient light mapping information reflecting the brightness characteristics corresponding to different entity factors in the current detection area. At the same time, through the analysis of the environmental image information set, the glare influence feature set reflecting the specific influence of each environmental glare influence factor on the indirect glare is determined. Through the combined analysis of the ambient light mapping information and the glare influence feature set, the indirect glare risk corresponding to each area in the detection area under the current light conditions is quantitatively analyzed, so as to construct the glare hidden danger distribution information. On this basis, the corresponding glare detection report is provided to the detection personnel, and the indirect glare generated under the influence of multi-dimensional factors in the detection area is accurately evaluated, so that the glare hidden dangers can be discovered in time, and the negative impact of glare on users can be significantly reduced.

[0039] Figure 1 A schematic diagram of an application scenario provided by the present application. In the process of lamp glare detection, the method provided by the present application is applied to accurately evaluate the indirect glare under the influence of multi-dimensional factors in the detection area.

[0040] Specifically, the method of the present application is applied to any server, which communicates with an imaging brightness meter and a visual sensor respectively. The server obtains and analyzes the brightness distribution information provided by the imaging brightness meter and the environmental image information set provided by the visual sensor, determines the ambient light mapping information reflecting the brightness characteristics corresponding to different entity factors in the current detection area, and determines the glare influence feature set reflecting the specific influence of each environmental glare influence factor on the indirect glare through the analysis of the environmental image information set. Through the combined analysis of the ambient light mapping information and the glare influence feature set, the indirect glare risk corresponding to each area in the detection area under the current light conditions is quantitatively analyzed, thereby constructing the glare hidden danger distribution information. On this basis, the corresponding glare detection report is provided to the detection personnel, and the indirect glare generated under the influence of multi-dimensional factors in the detection area is accurately evaluated, so that the glare hidden danger is discovered in time, and the negative impact of glare on users is significantly reduced. The specific implementation method can refer to the following embodiments.

[0041] Figure 2 This is a flow chart of an intelligent detection method for a lamp provided in an embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. Figure 2As shown, the method includes: S201, obtaining brightness distribution information and an environment image information set, analyzing the brightness distribution information based on the environment image information set, and determining environment light mapping information.

[0042] The brightness distribution information may be light intensity data affected by the target detection lamps at different positions in the current detection area, and the brightness distribution information may be obtained by an imaging brightness meter.

[0043] The environmental image information set may be a set of image information of different perspectives corresponding to the current detection area, and the environmental image information set may be obtained by visual sensors arranged at different positions within the detection area.

[0044] The ambient light mapping information may be information obtained by combining the brightness distribution information with the ambient image and used to characterize the light distribution in the real space.

[0045] Specifically, the core of lamp glare detection is to identify and evaluate the impact of light sources on visual comfort. The purpose is to ensure that the lighting conditions in the environment can effectively support human visual activities without causing discomfort or visual fatigue. Glare types are divided into direct glare and indirect glare. Direct glare is usually caused by the light projected by the light source directly causing glare to the human eye. Direct glare depends on the light intensity of the lamp. The brightness distribution information obtained by the imaging brightness meter can accurately measure the light intensity at each position in the detection area under the influence of the lamp, and then make an accurate assessment of direct glare. Indirect glare refers to the glare effect of light on the human eye caused by reflection or scattering. Indirect glare usually does not come from the direct light source, but from the surrounding The impact of domestic objects on light sources is relatively hidden from the human eye. Since the brightness distribution information does not contain information on the various physical factors in the environment that affect indirect glare, indirect glare is difficult to analyze directly based on the brightness distribution information. It is necessary to combine the different physical factors in the environment that affect indirect glare, and use mathematical analysis and image analysis technology to identify the different physical factors in the environmental image information set that affect indirect glare. In combination with the brightness distribution information, the brightness characteristics of each physical factor currently reflected in the brightness distribution information are mapped and processed to obtain the ambient light mapping information, which provides a scientific data framework for subsequent analysis of indirect glare.

[0046] S202: Analyze the environmental image information set, analyze and extract environmental glare influencing factors, and determine a glare influencing feature set.

[0047] The environmental glare influencing factor may be a physical factor that has an impact on the indirect glare in the current detection area.

[0048] The glare impact feature set may be a set of feature parameters used to characterize the specific impact of physical factors on the existence of indirect glare, such as surface reflectivity, color, etc.

[0049] Specifically, since indirect glare is caused by the influence of physical factors in the detection area, it is manifested as the light generated by the lamp light source being reflected or scattered by the physical objects in the detection area, causing glare to the human eye. The intensity of the reflection or scattering effect of the physical objects on the light is highly correlated with the surface reflectivity, geometric structure and color of the objects. Before measuring the specific impact of the physical objects on the indirect glare, the environmental image information set is analyzed through mathematical analysis combined with image analysis technology to identify the environmental glare influencing factors, and extract the glare influence features from the environmental glare influencing factors to construct a glare influence feature set, providing a data basis for the subsequent feature fusion analysis of indirect glare.

[0050] S203: Analyze the glare impact feature set based on the ambient light mapping information to determine the glare hazard distribution information.

[0051] The glare hazard distribution information may be information constructed by comprehensively analyzing the glare gain risks corresponding to all physical factors that affect indirect glare in the detection area and used to characterize the corresponding glare risks of each area in the detection area.

[0052] Specifically, after obtaining the ambient light mapping information and the glare impact feature set respectively, mathematical analysis is used to quantify the gain impact of each impact feature in the glare impact feature set on the generation of indirect glare under the light conditions reflected by the current ambient light mapping information, and then determine the glare hazards existing in each local area in the detection area under the gain impact of each entity factor on the indirect glare, so as to construct the glare hazard distribution information, so that the detection personnel can intuitively understand the glare risks existing in the current detection area.

[0053] S204: Determine and output a glare detection report based on the glare hazard distribution information.

[0054] The glare detection report may be report information including detection results of indirect glare risk in each local area within the current detection area.

[0055] Specifically, through data visualization technology, the glare hazard distribution information is visualized and integrated to intuitively display the distribution of indirect glare hazards in the current detection area, and through human-computer interaction equipment, such as high-definition display screens, the visualization data corresponding to the glare hazard distribution information is provided to the detection personnel.

[0056] Through this solution, the environmental image information set and the brightness distribution information are analyzed to determine the environmental light mapping information reflecting the brightness characteristics corresponding to different entity factors in the current detection area. At the same time, through the analysis of the environmental image information set, the glare influence feature set reflecting the specific influence of each environmental glare influence factor on the indirect glare is determined. Through the combined analysis of the environmental light mapping information and the glare influence feature set, the indirect glare risk corresponding to each area in the detection area under the current light conditions is quantitatively analyzed to construct the glare hidden danger distribution information. On this basis, the corresponding glare detection report is provided to the detection personnel to accurately evaluate the indirect glare generated by the multi-dimensional factors in the detection area, so that the glare hidden dangers can be discovered in time and the negative impact of glare on users can be significantly reduced.

[0057] In some embodiments, based on the brightness distribution information, a number of brightness intervals are extracted; based on the environmental image information set, the brightness distribution information is spatially mapped to determine the brightness mark image information; based on the brightness mark image information, the environmental image information set is brightness layered according to the number of brightness intervals to determine a number of brightness layered areas; based on the brightness distribution information, a number of brightness layered areas are analyzed to determine the regional average brightness and the number of adjacent layered areas corresponding to each brightness layered area; based on the regional average brightness and the number of adjacent layered areas, the illumination transition mutation index of each brightness layered area is determined; the illumination transition mutation index of each brightness layered area is used as the ambient light mapping information.

[0058] The brightness interval may be a light intensity interval of different ranges existing in the brightness distribution information.

[0059] The spatial coordinate mapping process may be a process of mapping the brightness value in the brightness distribution information with the spatial position in the environment image.

[0060] The brightness-marked image information may be image information with brightness marks generated after marking corresponding brightness values ​​of different regions in the environment image in the brightness distribution information.

[0061] The brightness layered region may be a region used to represent different brightness levels within an image.

[0062] The regional average brightness may be the arithmetic mean of the brightness corresponding to the brightness layered area.

[0063] The number of adjacent hierarchical regions may be the total number of other brightness hierarchical regions adjacent to the current brightness hierarchical region.

[0064] The illumination transition mutation index may be quantitative data for characterizing light non-uniformity in a light transition region between one brightness layered region and other brightness layered regions.

[0065] Specifically, the brightness distribution information reflects the overall brightness distribution of light in the current detection area. The data level is high and the data granularity is large, so it is difficult to provide targeted data support for the analysis of indirect glare. By presetting the brightness interval division rules or according to the current light source parameters, the typical brightness interval in the brightness distribution information is divided to obtain several brightness intervals, which can more clearly analyze the brightness changes and possible glare areas. At the same time, by mapping the brightness distribution values ​​in the detection area reflected in the brightness distribution information to the corresponding spatial positions in the environmental image set, the brightness values ​​are made to correspond one to one with the spatial coordinate positions in the image, and the brightness mark image information is obtained to characterize the correlation between the brightness distribution information and the environmental image, so as to clarify the actual mapping of the abstract data in the brightness distribution information in the environmental image, and provide a basis for subsequent glare analysis. On this basis, according to the corresponding brightness value ranges in several brightness intervals, the brightness mark image information is analyzed, and the corresponding areas of the brightness intervals in the environmental image are divided to obtain several brightness layered areas, which further provide an intuitive description for the distribution of light in the image.

[0066] In addition to the reflection and scattering of the objects mentioned in the above embodiments, the generation of indirect glare is also affected by the local difference of light. It takes time for the human eye to adapt to light and there is a lag. When the local light difference is too large, even if the absolute light intensity does not reach the glare standard, it will still cause glare to the human visual sense. Similarly, the reflection and scattering of objects can also be regarded as the sudden non-smooth intervention of light, which causes the human eye to feel glare. It can be seen that whether it is the indirect glare caused by light reflection or the indirect glare caused by local light difference, it is affected by the smoothness of light change. The less smooth the light change, the greater the risk of indirect glare. Therefore, after determining a number of brightness stratified areas, the average brightness of the area corresponding to each brightness stratified area and the number of adjacent stratified areas are extracted. On this basis, through mathematical analysis, the illumination transition mutation index used to describe whether the light transitions smoothly between different brightness stratified areas is quantified. The larger the illumination transition mutation index, the greater the degree of mutation of the light in the current transition process, the lower the light transition smoothness, and the greater the corresponding indirect glare risk. This can be used as ambient light mapping information to provide scientific data guidance for subsequent analysis of indirect glare.

[0067] Through this scheme, a number of brightness intervals are divided according to the brightness distribution information, and at the same time, the brightness distribution information is subjected to spatial coordinate mapping processing according to the environmental image information set to determine the brightness mark image information. On this basis, a number of brightness layered areas are divided, and according to the average brightness of the area corresponding to each of the brightness layered areas and the number of adjacent layered areas, the illumination transition mutation index used to reflect the degree of light transition mutation between different brightness layered areas is determined, so as to construct the ambient light mapping information, fully considering the common characteristics of light under different indirect glare conditions, so that the subsequent indirect glare analysis process based on the ambient light mapping information is more accurate and comprehensive.

[0068] In some embodiments, the illumination transition mutation index of each brightness layered region is determined according to the average brightness of the region and the number of adjacent layered regions, specifically, the following formula (1): (1) in, For the The illumination transition mutation index of each brightness layered area, For the The average brightness of the brightness layered area, For the The number of adjacent stratified regions of a brightness stratified region, For the The brightness layered area adjacent to the The regional average brightness of the brightness layered area.

[0069] Specifically, through formula (1) Describes the average relative deviation between the average brightness of light in all adjacent layered areas adjacent to the current brightness layered area and the average brightness of light in the current brightness layered area. The larger the average relative deviation, the less smooth the process of light transitioning from the current brightness layered area to other brightness layered areas, and the higher the corresponding illumination transition mutation index. On this basis, the introduction As a benchmark factor, the illumination transition mutation index is normalized, and then the illumination transition mutation index is quantified to reflect whether the light transition is smooth.

[0070] Through this scheme, mathematical analysis methods are used to accurately quantify the illumination transition mutation index used to reflect whether the light transition is smooth based on the average regional brightness corresponding to each brightness stratified area and the number of adjacent stratified areas through scientific mathematical formulas, thereby improving the accuracy and scientificity of the illumination transition mutation index evaluation process, thereby improving the accuracy and analysis efficiency of the subsequent indirect glare analysis process based on the illumination transition mutation index.

[0071] In some embodiments, according to a preset denoising and normalization strategy, each environmental image in the environmental image information set is uniformly denoised and normalized to determine a preprocessed image information set; based on the preprocessed image information set, entity recognition and edge contour extraction are performed on objects in each preprocessed image to determine an object surface feature set and an object geometric feature information set; based on the preprocessed image information set, a two-stage conversion is performed on the color space corresponding to each preprocessed image to determine the human eye adaptive color features of each object and construct an object color feature set; based on the object surface feature set, the object geometric feature information set and the color feature set, a glare impact feature set is constructed.

[0072] The preset denoising and normalization strategy can be a preset strategy for performing noise filtering and pixel normalization processing on all images in the environmental image information set. The preset denoising and normalization strategy can use a computer vision library, such as OpenCV (Open Source Computer Vision Library), to perform Gaussian blur and linear normalization processing on the image.

[0073] The preprocessed image information set may be a set consisting of all denoising and normalized environment images.

[0074] The object surface feature set may be a feature set used to characterize the surface properties of an object, such as surface roughness.

[0075] The object geometric feature information set may be a feature set used to characterize the geometric structure of an object.

[0076] The human eye adaptive color feature may be a feature used to characterize the color of an object based on the human eye's color perception characteristics.

[0077] The object color feature set may be a set of color features corresponding to human eye adaptation of all objects.

[0078] Specifically, in the process of identifying and analyzing different environmental glare influencing factors in the environmental image information set, a unified denoising and normalization process is performed on each environmental image through a preset denoising and normalization strategy to determine the preprocessed image information set. Denoising improves the accuracy of subsequent image analysis and reduces the possibility of misidentifying objects. Normalization ensures the consistency of images under different shooting conditions, which facilitates comparison and feature extraction. On this basis, based on a deep learning model and combined with an edge detection algorithm in image processing technology, entity recognition and edge contour extraction of objects in each preprocessed image are realized. The surface feature set of the object is determined according to the entity recognition results, and the edge contour extraction is performed according to the edge contour extraction results. The edge contour extraction result is used to determine the object geometric feature information set. At the same time, in order to adapt to the display requirements of the display device, the image usually adopts the RGB color space. In the glare detection process, since glare mainly affects the perception of the human eye, the default RGB color space of the image is converted into a color space suitable for describing the human eye perception characteristics through a two-stage color space conversion process, so that the obtained object color feature set fits the human eye perception state, and the accuracy of subsequent glare detection is improved. The above-mentioned object surface feature set, object geometric feature information set and color feature set are used as the glare influence feature set to comprehensively reflect the impact of environmental glare factors on indirect glare.

[0079] Through this scheme, the environmental image information set is denoised and normalized to determine the preprocessed image information set. On this basis, the object surface feature set and the object geometric feature information set are determined by entity recognition and edge contour extraction of the objects in the preprocessed image. At the same time, the color space of the preprocessed image is converted in two stages so that the object color feature set is consistent with the human eye perception state. The above-mentioned object surface feature set, object geometric feature information set and color feature set are used as the glare influence feature set to comprehensively reflect the impact of environmental glare influencing factors on indirect glare.

[0080] In some embodiments, based on the pre-processed image information set, a two-stage conversion is performed on the color space corresponding to each pre-processed image to determine the human eye adaptive color feature of each object, specifically, the following formula (2): (2) in, is the human eye adaptive color feature corresponding to the current object pixel, is the RGB color space value corresponding to each pixel value of the current object, It is the function of converting RGB color space to XYZ color space. Converts XYZ color space to Lab color space function.

[0081] The RGB color space to XYZ color space function may be a mathematical function used to convert the RGB color space to the XYZ color space, and the RGB color space to XYZ color space function may use a linear transformation algorithm.

[0082] The XYZ color space to Lab color space conversion function may be a mathematical function used to convert the XYZ color space to the Lab color space. The XYZ color space to Lab color space conversion function may adopt a CIE76 Lab conversion algorithm.

[0083] Specifically, the default RGB color space of the ambient image is converted into the corresponding XYZ color space through formula (2), and the obtained XYZ color space is further converted into the corresponding Lab color space. The Lab color space is a color space based on human visual perception. L represents brightness, and a and b are responsible for separating color information. This separation makes color analysis more consistent with human visual experience. The Lab color space performs well in processing dark and bright color details, and can assist in the subsequent analysis of indirect glare risks, so that the glare detection results are consistent with the perception characteristics of the human eye.

[0084] Through this solution, based on the preprocessed image information set, a two-stage conversion is performed on the color space corresponding to each preprocessed image, and the default RGB color space of the environmental image is converted into the Lab color space that conforms to the human visual experience, so as to assist in the subsequent analysis of indirect glare risks, make the glare detection results conform to the human eye perception characteristics, and improve the accuracy and scientificity of subsequent glare detection results.

[0085] In some embodiments, the illumination transition mutation index of each brightness layered area in the ambient light mapping information is compared with a preset illumination mutation threshold, and all corresponding brightness layered areas with illumination transition mutation index greater than the preset illumination mutation threshold are extracted to determine a number of key analysis areas; based on the several key analysis areas, the glare impact feature set is screened to extract the object surface feature set, object geometric feature information set and color feature set of each object in the key analysis area; based on the object surface feature set, the surface roughness of each object is determined; the object geometric feature information set is analyzed to extract the edge sharpness of each object; based on the surface roughness, edge sharpness and color feature set, the surface reflection influence factor, edge sharpness influence factor and color gain influence factor are respectively determined; based on the brightness distribution information, the average background brightness is determined, and based on the average background brightness, the glare risk gain value corresponding to each object is determined according to the surface reflection influence factor, edge sharpness influence factor and color gain influence factor, so as to construct the glare hazard distribution information.

[0086] The preset illumination mutation threshold may be a preset benchmark value for measuring whether the illumination transition mutation index will cause a risk of indirect glare. The preset illumination mutation threshold may be obtained by fitting simulated detection data.

[0087] The key analysis area may be the corresponding brightness layered area where there is a risk of inducing indirect glare.

[0088] Surface roughness can be a quantitative value used to characterize the roughness of the surface of an object.

[0089] Corner sharpness may be a quantitative value used to characterize the sharpness of the corners of an object.

[0090] The surface reflection impact factor can be a quantitative factor used to characterize the impact of the light reflection characteristics of the object surface on the risk of indirect glare.

[0091] The edge sharpness impact factor may be a quantitative factor used to characterize the impact of the edge sharpness of an object on the risk of indirect glare.

[0092] The color gain impact factor can be a quantitative factor that characterizes the impact of the color characteristics of an object on the risk of indirect glare.

[0093] The average background brightness may be a value used to characterize the background light intensity of an image.

[0094] The glare risk gain value may be a quantitative value used to characterize the degree of gain in the indirect glare risk under the influence of the current object.

[0095] Specifically, since the numerical value of the illumination transition mutation index is positively correlated with the risk of indirect glare, by comparing the illumination transition mutation index of each brightness layered area with the preset illumination mutation threshold, all corresponding brightness layered areas with illumination transition mutation index greater than the preset illumination mutation threshold are divided into key analysis areas, so as to provide accurate area screening conditions for indirect glare risk analysis and improve analysis efficiency. On this basis, the surface feature set, geometric feature information set and color feature set of each object in the key analysis area are extracted, and the surface roughness and edge sharpness corresponding to the object are determined according to the surface feature set and the geometric feature information set of the object through regression algorithm and edge detection algorithm respectively. The surface reflection influencing factor, edge sharpness influencing factor and color gain influencing factor are quantified according to the surface roughness, edge sharpness and color feature set respectively through mathematical analysis. The surface reflection influencing factor reflects the light reflection on the surface of the object. The reflective characteristics have a greater impact on the gain of indirect glare. The smoother the surface, the higher the reflectivity. Highly reflective surfaces (such as mirrored metal or smooth plastic) will increase the intensity of reflected light in the environment, thereby increasing the probability of indirect glare. The edge sharpness influencing factor reflects the gain effect of the edge characteristics of the object on indirect glare. The sharp edges of the object will cause strong reflection of light, increasing the intensity of light. This light will produce a more obvious highlight area at the edge of the object and enhance the contrast, thereby causing glare. The color gain influencing factor reflects the gain effect of the color characteristics of the object on indirect glare. The color of the object will affect the absorption and reflection characteristics of light. For example, light-colored objects (such as white or light yellow) usually reflect more light, while dark-colored objects (such as black or dark blue) absorb light relatively. Highly reflective light-colored objects will increase the risk of indirect glare, and some bright colors (such as fluorescent colors) are prone to cause visual fatigue and discomfort, thereby accelerating the generation of glare.

[0096] Through the image segmentation algorithm in image processing technology, the foreground area and background area in the environmental image are divided, and the average background brightness corresponding to the background area is extracted according to the brightness distribution information. The average background brightness is helpful for quantitative analysis of the degree of local light difference. Through further mathematical analysis, based on the average background brightness, the surface reflection influencing factor, the edge sharpness influencing factor and the color gain influencing factor are comprehensively analyzed to quantify the glare risk gain value corresponding to each object, so as to construct the glare hazard distribution information and realize the detection of indirect glare hazards.

[0097] Through this scheme, according to the comparison result of the illumination transition mutation index and the preset illumination mutation threshold, several key analysis areas are screened, and the surface feature set, geometric feature information set and color feature set of each object in the key analysis area are extracted. On this basis, the surface roughness and edge sharpness corresponding to the object are analyzed, and then the surface reflection influencing factor, edge sharpness influencing factor and color gain influencing factor are determined. In combination with the average background brightness, the glare risk gain value of each object is evaluated, so as to construct the glare hazard distribution information, so that the analysis process of indirect glare hazards is closely combined with the characteristics of each object in the environment, improve the accuracy and pertinence of the glare hazard distribution information, and significantly improve the accuracy of indirect glare detection under the influence of multi-dimensional factors.

[0098] In some embodiments, based on the average background brightness, according to the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor, the glare risk gain value corresponding to each object is determined, specifically, as follows: (3) in, is the glare risk gain value, is the surface reflection factor, is the factor affecting the sharpness of the edges, is the color gain influencing factor, is the average background brightness, is the preset weighted impact index, The preset sensitivity adjustment index.

[0099] The preset weighted influence index may be a quantitative value used to characterize the nonlinear influence of the average background brightness on the glare risk gain value, and the preset weighted influence index may be obtained by fitting experimental data.

[0100] The preset sensitivity adjustment index may be a quantitative value used to characterize the sensitivity of the human eye to adapt to ambient light. The preset sensitivity adjustment index may be obtained by fitting experimental data with authoritative paper data.

[0101] Specifically, through formula (3) Reflecting the cumulative effect of surface reflection, corner sharpness and color on glare, the combination of the three can be regarded as the "potential to cause indirect glare" of the object itself; The effect of average background brightness is amplified to increase the sensitivity of the formula to indirect glare caused by large differences in light; It reflects the regulatory effect of average background brightness on indirect glare risk. As background brightness increases, the indirect glare risk decreases relatively (due to human eye adaptation and reduced light differences). The glare risk gain value corresponding to each object is then quantified.

[0102] Through this scheme, mathematical analysis is used, based on the average background brightness, according to the surface reflection influence factor, edge sharpness influence factor and color gain influence factor, through a clear mathematical relationship language, to describe the influence of the above parameters on indirect glare, so as to quantify the glare risk gain value corresponding to each object, significantly improving the scientificity and accuracy of the glare risk gain value.

[0103] In some embodiments, according to the surface roughness, the edge sharpness and the color feature set, the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor are determined respectively, specifically as follows: (4) in, is the surface reflection factor, is the preset slope factor, is the surface roughness, is the preset roughness reference value, is the factor affecting the sharpness of the edges, is the edge sharpness, To preset the sharpness sensitivity index, is the color gain influencing factor, is the human eye adaptive color feature corresponding to the current object pixel, is the preset color gain relationship coefficient.

[0104] The preset slope factor may be a mathematical factor used to characterize the degree of influence of the change in the roughness of the surface of an object on its light reflectivity, and the preset slope factor may be obtained by fitting experimental data.

[0105] The preset roughness reference value may be a reference value used to characterize the direct difference between the surface roughness of an object and the ideal surface roughness. The preset roughness reference value may be obtained by fitting experimental data.

[0106] The preset sharpness sensitivity index may be a quantitative value used to characterize the influence relationship between edge sharpness and indirect glare risk, and the preset sharpness sensitivity index may be obtained by fitting experimental data.

[0107] The preset color gain relationship coefficient may be a quantitative value used to characterize the degree of light gain of the object color, and the preset color gain relationship coefficient may be obtained by fitting experimental data.

[0108] Specifically, through formula (4) Quantify the factors affecting surface reflection. The surface reflection ability of an object is usually inversely correlated with the roughness of the object. The smaller the roughness, the stronger the reflection ability. The sigmoid function is used here to describe the nonlinear response of the surface reflection ability. The slope of the function is limited by the preset slope factor. Quantify the influencing factors of corner sharpness. Corner sharpness has a significant impact on the degree of light reflection. Generally speaking, the higher the corner sharpness, the more severe the glare. Quantify the color gain influencing factors. The vividness and brightness of the color of an object will directly affect the human eye's perception of glare. Generally speaking, the brighter the color, the more obvious the glare gain effect. In the Lab color space, Indicates the brightness of the surface of an object. It can be used to estimate the comprehensive saturation of color; through the above mathematical analysis process, the scientific quantification of surface reflection influencing factors, edge sharpness influencing factors and color gain influencing factors can be achieved.

[0109] Through this scheme, mathematical analysis is used to scientifically quantify the surface reflection influence factor, edge sharpness influence factor and color gain influence factor according to the surface roughness, edge sharpness and color feature set, thereby significantly improving the accuracy and scientificity of the surface reflection influence factor, edge sharpness influence factor and color gain influence factor, thereby improving the accuracy of the assessment of indirect glare risk based on the above factors.

[0110] In some embodiments, according to the glare risk gain value corresponding to each object in the glare hazard distribution information, several brightness layered areas are analyzed to determine the comprehensive glare risk corresponding to each brightness layered area; according to the comprehensive glare risk corresponding to each brightness layered area, the corresponding area in each environmental image in the environmental image information set is marked with glare risk to determine the glare risk distribution image set; the glare risk distribution image set is visualized and output as a glare detection report.

[0111] The comprehensive glare risk may be a quantitative value determined after comprehensive consideration of the glare risk gain values ​​corresponding to different objects in the brightness stratification area, and is used to characterize the overall indirect glare risk in the brightness stratification area.

[0112] The glare risk distribution image set may be a set of image information generated after quantitatively marking the comprehensive glare risk of each area in each environmental image.

[0113] Specifically, after determining the glare hazard distribution information, linear summation or weighted summation is performed on the glare risk gain values ​​corresponding to all objects in the corresponding brightness layered area according to the objects contained in each brightness layered area, and the comprehensive glare risk corresponding to each brightness layered area is determined. The comprehensive glare risk reflects the overall indirect glare risk in the corresponding brightness layered area. On this basis, the areas in the environmental image information set corresponding to the brightness layered areas are marked with glare risks, so as to determine a glare risk distribution image set that can intuitively reflect the overall indirect glare risk of each area in the environment, and the glare risk distribution image set is visualized and output as a glare detection report to improve the readability of the glare detection report.

[0114] Through this solution, based on the glare hazard distribution information, the comprehensive glare risk corresponding to each of the brightness stratified areas is analyzed and obtained, and the corresponding area in each environmental image in the environmental image information set is marked with glare risk, so as to obtain a glare risk distribution image set that can intuitively reflect the overall indirect glare risk of each area in the environment, and the glare risk distribution image set is visualized and output as a glare detection report to improve the readability of the glare detection report.

[0115] Figure 3 A schematic diagram of the structure of an intelligent detection system for a lamp provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, a smart detection system 300 for lamps in this embodiment includes: a mapping analysis module 301 , a feature extraction module 302 , a hidden danger analysis module 303 and an output module 304 .

[0116] A mapping analysis module 301 is used to obtain brightness distribution information and an environmental image information set, analyze the brightness distribution information based on the environmental image information set, and determine ambient light mapping information; A feature extraction module 302 is used to analyze the environmental image information set based on the brightness distribution information, analyze and extract environmental glare influencing factors, and determine a glare influencing feature set; A hidden danger analysis module 303 is used to analyze the glare impact feature set based on the ambient light mapping information to determine glare hidden danger distribution information; The output module 304 is used to determine and output a glare detection report according to the glare hidden danger distribution information.

[0117] Optionally, the mapping analysis module 301 is specifically used to: extract a number of brightness intervals according to the brightness distribution information; perform spatial coordinate mapping processing on the brightness distribution information according to the environmental image information set to determine brightness mark image information; based on the brightness mark image information, perform brightness stratification processing on the environmental image information set according to the several brightness intervals to determine a number of brightness stratification areas; based on the brightness distribution information, analyze the several brightness stratification areas to determine the regional average brightness and the number of adjacent stratification areas corresponding to each brightness stratification area; determine the illumination transition mutation index of each brightness stratification area according to the regional average brightness and the number of adjacent stratification areas; and use the illumination transition mutation index of each brightness stratification area as the ambient light mapping information.

[0118] Optionally, when the mapping analysis module 301 determines the illumination transition mutation index of each brightness layered region according to the average brightness of the region and the number of adjacent layered regions, the specific formula is as follows: ;in, For the The illumination transition mutation index of the brightness layered area, For the The average brightness of the brightness layered areas, For the The number of adjacent layered regions of the brightness layered region, For the The brightness layered area is adjacent to the The average brightness of the brightness layered areas.

[0119] Optionally, the feature extraction module 302 is specifically used to: perform unified denoising and normalization processing on each environmental image in the environmental image information set according to a preset denoising and normalization strategy to determine a preprocessed image information set; based on the preprocessed image information set, perform entity recognition and edge contour extraction on objects in each preprocessed image to determine an object surface feature set and an object geometric feature information set; based on the preprocessed image information set, perform a two-stage conversion on the color space corresponding to each preprocessed image to determine the human eye adaptive color features of each object and construct an object color feature set; and construct the glare impact feature set based on the object surface feature set, the object geometric feature information set and the color feature set.

[0120] Optionally, when the feature extraction module 302 performs a two-stage conversion on the color space corresponding to each preprocessed image based on the preprocessed image information set to determine the human eye adaptive color feature of each object, the specific formula is as follows: ;in, is the human eye adaptive color feature corresponding to the current object pixel, is the RGB color space value corresponding to each pixel value of the current object, It is the function of converting RGB color space to XYZ color space. Converts XYZ color space to Lab color space function.

[0121] Optionally, the hidden danger analysis module 303 is specifically used to: compare the illumination transition mutation index of each brightness layered area in the ambient light mapping information with a preset illumination mutation threshold, extract all corresponding brightness layered areas whose illumination transition mutation index is greater than the preset illumination mutation threshold, and determine a number of key analysis areas; screen the glare impact feature set according to the several key analysis areas, extract the object surface feature set, the object geometric feature information set and the color feature set of each object in the key analysis area; determine the surface roughness of each object according to the object surface feature set; analyze the object geometric feature information set to extract the edge sharpness of each object; determine the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor according to the surface roughness, the edge sharpness and the color feature set; determine the average background brightness according to the brightness distribution information, and based on the average background brightness, determine the glare risk gain value corresponding to each object according to the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor, so as to construct the glare hidden danger distribution information.

[0122] Optionally, when the hidden danger analysis module 303 determines the glare risk gain value corresponding to each object based on the average background brightness, the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor, the specific formula is as follows: ;in, is the glare risk gain value, is the surface reflection influence factor, is the edge sharpness influencing factor, is the color gain influencing factor, is the average background brightness, is the preset weighted impact index, The preset sensitivity adjustment index.

[0123] Optionally, when the hidden danger analysis module 303 determines the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor according to the surface roughness, the edge sharpness and the color feature set, the specific formulas are as follows: ; in, is the surface reflection influence factor, is the preset slope factor, is the surface roughness, is the preset roughness reference value, is the edge sharpness influencing factor, is the edge sharpness, To preset the sharpness sensitivity index, is the color gain influencing factor, is the human eye adaptive color feature corresponding to the current object pixel, is the preset color gain relationship coefficient.

[0124] Optionally, the output module 304 is specifically used to: analyze a number of the brightness stratified areas according to the glare risk gain value corresponding to each object in the glare hazard distribution information, and determine the comprehensive glare risk corresponding to each of the brightness stratified areas; mark the corresponding area in each environmental image in the environmental image information set with glare risk according to the comprehensive glare risk corresponding to each of the brightness stratified areas, and determine a glare risk distribution image set; and visualize and output the glare risk distribution image set as the glare detection report.

[0125] The system of this embodiment can be used to execute the method of any of the above embodiments. The implementation principles and technical effects are similar and will not be described in detail here.

Claims

1. An intelligent detection method for lamps, characterized in that: include: Acquire brightness distribution information and an environment image information set, analyze the brightness distribution information based on the environment image information set, and determine the environment light mapping information; Analyze the environmental image information set, analyze and extract the environmental glare influencing factors, and determine the glare influencing feature set; Based on the ambient light mapping information, analyzing the glare impact feature set to determine glare hazard distribution information; A glare detection report is determined and outputted according to the glare hazard distribution information.

2. The method according to claim 1, characterized in that The step of analyzing the brightness distribution information based on the environment image information set to determine the environment light mapping information includes: Extracting a plurality of brightness intervals according to the brightness distribution information; According to the environmental image information set, the brightness distribution information is subjected to spatial coordinate mapping processing to determine brightness mark image information; Based on the brightness marked image information, according to the plurality of brightness intervals, performing brightness layering processing on the environment image information set to determine a plurality of brightness layering areas; Based on the brightness distribution information, analyzing the plurality of brightness layered regions, and determining the average brightness of each brightness layered region and the number of adjacent layered regions; Determine the illumination transition mutation index of each brightness layered region according to the average brightness of the region and the number of adjacent layered regions; The illumination transition mutation index of each brightness layered area is used as the ambient light mapping information.

3. The method according to claim 2, characterized in that The illumination transition mutation index of each brightness layered region is determined according to the average brightness of the region and the number of adjacent layered regions, specifically as follows: ； in, For the The illumination transition mutation index of the brightness layered area, For the The average brightness of the brightness layered areas, For the The number of adjacent layered regions of the brightness layered region, For the The brightness layered area is adjacent to the The average brightness of the brightness layered areas.

4. The method according to claim 2, characterized in that: The step of analyzing the environmental image information set, analyzing and extracting environmental glare influencing factors, and determining a glare influencing feature set includes: According to a preset denoising and normalization strategy, each environmental image in the environmental image information set is subjected to a unified denoising and normalization process to determine a preprocessed image information set; Based on the preprocessed image information set, entity recognition and edge contour extraction are performed on the objects in each preprocessed image to determine the object surface feature set and the object geometric feature information set; Based on the preprocessed image information set, a two-stage conversion is performed on the color space corresponding to each preprocessed image, so as to determine the human eye adaptive color feature of each object and construct an object color feature set; The glare impact feature set is constructed according to the object surface feature set, the object geometric feature information set and the color feature set.

5. The method according to claim 4, characterized in that Based on the pre-processed image information set, the color space corresponding to each pre-processed image is converted in two stages to determine the human eye adaptive color feature of each object, which is specifically the following formula: ； in, is the human eye adaptive color feature corresponding to the current object pixel, is the RGB color space value corresponding to each pixel value of the current object, It is the function of converting RGB color space to XYZ color space. Converts XYZ color space to Lab color space function.

6. The method according to claim 5, characterized in that The step of analyzing the glare impact feature set based on the ambient light mapping information to determine glare hazard distribution information includes: Compare the illumination transition mutation index of each brightness layered area in the ambient light mapping information with a preset illumination mutation threshold, extract all corresponding brightness layered areas whose illumination transition mutation index is greater than the preset illumination mutation threshold, and determine a number of key analysis areas; According to the plurality of key analysis areas, the glare impact feature set is screened to extract the object surface feature set, the object geometric feature information set and the color feature set of each object in the key analysis area; Determining the surface roughness of each object based on the set of surface features of the object; Analyze the geometric feature information set of the objects to extract the edge and corner sharpness of each object; According to the surface roughness, the edge sharpness and the color feature set, a surface reflection influence factor, an edge sharpness influence factor and a color gain influence factor are determined respectively; According to the brightness distribution information, the average background brightness is determined, and based on the average background brightness, according to the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor, the glare risk gain value corresponding to each object is determined, so as to construct the glare hazard distribution information.

7. The method according to claim 6, characterized in that Based on the average background brightness, the glare risk gain value corresponding to each object is determined according to the surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor, which is specifically the following formula: ； in, is the glare risk gain value, is the surface reflection influence factor, is the edge sharpness influencing factor, is the color gain influencing factor, is the average background brightness, is the preset weighted impact index, The preset sensitivity adjustment index.

8. The method according to claim 7, characterized in that The surface reflection influence factor, the edge sharpness influence factor and the color gain influence factor are determined according to the surface roughness, the edge sharpness and the color feature set, respectively, which are specifically the following formulas: ； in, is the surface reflection influence factor, is the preset slope factor, is the surface roughness, is the preset roughness reference value, is the edge sharpness influencing factor, is the edge sharpness, To preset the sharpness sensitivity index, is the color gain influencing factor, is the human eye adaptive color feature corresponding to the current object pixel, is the preset color gain relationship coefficient.

9. The method according to claim 8, characterized in that The step of determining and outputting a glare detection report according to the glare hidden danger distribution information includes: Analyze a plurality of brightness stratified areas according to the glare risk gain value corresponding to each object in the glare hazard distribution information, and determine the comprehensive glare risk corresponding to each brightness stratified area; According to the comprehensive glare risk corresponding to each brightness layered area, a corresponding area in each environmental image in the environmental image information set is marked with a glare risk mark to determine a glare risk distribution image set; The glare risk distribution image set is visualized and output as the glare detection report.

10. An intelligent detection system for lamps, characterized in that: include: A mapping analysis module, used to obtain brightness distribution information and an environmental image information set, analyze the brightness distribution information based on the environmental image information set, and determine the ambient light mapping information; A feature extraction module, used to analyze the environmental image information set based on the brightness distribution information, analyze and extract environmental glare influencing factors, and determine a glare influencing feature set; A hidden danger analysis module, used to analyze the glare impact feature set based on the ambient light mapping information and determine the glare hidden danger distribution information; The output module is used to determine and output a glare detection report according to the glare hidden danger distribution information.

Images