Building energy consumption control method and system using illumination image change

Abstract

The invention discloses a building energy consumption control method and system using illumination image change, and the method comprises the steps: obtaining indoor and outdoor illumination images of a building, generating a brightness distribution image, extracting a feature point set of an illumination change region, and correcting the feature point set in combination with the distribution information of a stable illumination region. And on the basis of the corrected feature point set, performing partition comparison on the illumination area, calculating a brightness difference index, and judging the uniformity of illumination distribution, thereby generating an energy-saving control or dynamic adjustment instruction, realizing refined energy consumption management on the illumination equipment and the sun-shading device, and improving the overall energy efficiency of the building.

Description

Technical Field

[0001] The present invention relates to the technical field of building energy consumption management, and specifically relates to a method and system for optimizing energy consumption control by using changes in illumination images. Background Art

[0002] With the development of building technology, the dynamic management and energy-saving control of building energy consumption have become the research focus. Existing energy consumption control systems mostly adjust based on sensor data such as temperature and humidity, and fail to fully utilize the illumination distribution information inside the building, resulting in limited regulation accuracy for lighting, shading and other devices. Some solutions attempt to introduce image acquisition technology to sense the illumination situation, but generally have problems such as simple image processing algorithms, inaccurate positioning of illumination change regions, and rough control logic, making it difficult to achieve refined optimization of energy consumption in complex illumination environments. Therefore, there is an urgent need for an intelligent control method that integrates image brightness analysis and regional difference judgment to improve the responsiveness and accuracy of building energy consumption regulation. Summary of the Invention

[0003] The purpose of the present invention is to provide a building energy consumption control method and system using changes in illumination images to solve the problems of low illumination perception accuracy, inaccurate regional recognition, and insensitive energy consumption regulation response in the prior art. By introducing a feature point correction mechanism and calculating the brightness difference index, the present invention can achieve refined analysis of the illumination distribution inside the building, and automatically generate energy-saving or adjustment instructions based on illumination uniformity, thereby improving the operation efficiency of lighting and shading devices and realizing dynamic optimization and intelligent management of building energy consumption.

[0004] Specifically, the present invention provides a building energy consumption control method using changes in illumination images, and the building energy consumption control method includes the following steps: Obtain indoor and outdoor illumination images of the building, and generate a brightness distribution image corresponding to the illumination image; Based on a preset illumination feature positioning method, obtain a set of feature points of the illumination change region in the brightness distribution image; Obtain the distribution information of the set of feature points to the stable illumination region in the brightness distribution image, and correct the set of feature points according to the distribution information to obtain a corrected set of feature points; Based on the corrected set of feature points, perform partition comparison on the illumination regions in the illumination image to obtain energy consumption optimization information for the illumination regions.

[0005] Further, the step of obtaining the distribution information of the set of feature points to the stable illumination region in the brightness distribution image includes: Set a first reference line and a second reference line that intersect at the center of the feature point set, pass through the stable illumination area and are perpendicular to each other, so that the first reference line intersects the illumination change area at a first intersection point and a second intersection point, and the second reference line intersects the illumination change area at a third intersection point and a fourth intersection point; Take the distances and brightness differences between the center of the feature point set and the first intersection point, the second intersection point, the third intersection point, and the fourth intersection point as distribution information.

[0006] Further, the step of correcting the feature point set according to the distribution information to obtain a corrected feature point set includes: Translate the center of the feature point set along the direction of the first reference line until the first distance and brightness difference between the center of the feature point set and the first intersection point and the second distance and brightness difference between the center of the feature point set and the second intersection point are equal to determine an intermediate feature point set; Translate the intermediate feature point set along the direction of the second reference line until the third distance and brightness difference between the intermediate feature point set and the third intersection point and the fourth distance and brightness difference between the intermediate feature point set and the fourth intersection point are equal to determine a corrected feature point set.

[0007] Further, the step of acquiring an indoor and outdoor illumination image of a building and generating a brightness distribution image corresponding to the illumination image includes: Collect an illumination image of the indoor and outdoor environment of the building and perform brightness normalization processing on the illumination image; Perform edge enhancement processing on the illumination image after brightness normalization processing to obtain the brightness distribution image.

[0008] Further, the step of obtaining a feature point set of an illumination change area in the brightness distribution image based on a preset illumination feature positioning method includes: Analyze the brightness distribution image based on a preset illumination feature positioning method; When a feature point set of an illumination change area in the brightness distribution image is recognized, take the feature point set as an initial feature point set and execute the step: obtain the distribution information of the feature point set to the stable illumination area in the brightness distribution image; When a feature point set of an illumination change area in the brightness distribution image cannot be recognized, output a corresponding prompt message.

[0009] Further, the step of partitioning and comparing illumination areas in the illumination image based on the corrected feature point set to obtain energy consumption optimization information of the illumination area includes: With the corrected feature point set as the center, divide the illumination area in the illumination image into a preset number of equal partitions; Obtain the brightness distribution information of the equal partitions, and determine the energy consumption optimization information of the illumination area according to the differences in the brightness distribution information between the equal partitions.

[0010] Further, the brightness distribution information includes brightness mean value and gradient value. The steps of obtaining the brightness distribution information of the equal partitions and determining the energy consumption optimization information of the illumination area according to the differences in the brightness distribution information between the equal partitions include: statistically calculating the brightness mean value and gradient value of each equal partition, and calculating the brightness difference index between the equal partitions. The brightness difference index is determined by the following formula: where, ΔL is the brightness difference index, M i is the brightness mean value of the i-th equal partition, M avg is the average brightness mean value of all equal partitions, G i is the brightness gradient value of the i-th equal partition, G avg is the average brightness gradient value of all equal partitions, D i is the distance between the center of the i-th equal partition and the building reference point, D max is the maximum building reference distance, N is the number of equal partitions; If the brightness difference index is within a preset range, the energy consumption optimization information is that the illumination area is uniform, and an energy-saving control instruction is generated; If the brightness difference index is not within the preset range, the energy consumption optimization information is that the illumination area is non-uniform, and a dynamic adjustment instruction is generated.

[0011] Further, the steps of generating the energy-saving control instruction or the dynamic adjustment instruction include: When the energy consumption optimization information is that the illumination area is uniform, generate an energy-saving control instruction to reduce the power of the lighting equipment or adjust the opening and closing angle of the sunshade device; When the energy consumption optimization information is that the illumination area is non-uniform, generate a dynamic adjustment instruction to increase the brightness of the lighting equipment or adjust the angle of the sunshade device for a specific partition.

[0012] The present invention also provides a building energy consumption control system using the change of illumination images. The building energy consumption control system includes: An image processing module, configured to obtain indoor and outdoor illumination images of a building and generate a brightness distribution image corresponding to the illumination images; A feature positioning module, configured to obtain a set of feature points of the illumination change area in the brightness distribution image based on a preset illumination feature positioning method; A feature correction module, configured to obtain the distribution information of the set of feature points to the stable illumination area in the brightness distribution image, and correct the set of feature points according to the distribution information to obtain a corrected set of feature points; A partition comparison module is used to perform partition comparison on the lighting areas in the lighting image based on the corrected feature point set, so as to obtain the energy consumption optimization information of the lighting areas.

[0013] Further, the partition comparison module further includes: A partition unit is used to divide the lighting areas in the lighting image into a preset number of equal partitions with the corrected feature point set as the center; An analysis unit is used to obtain the brightness distribution information of the equal partitions, and determine the energy consumption optimization information of the lighting areas according to the differences in the brightness distribution information between the equal partitions; An instruction generation unit is used to generate an energy-saving control instruction or a dynamic adjustment instruction according to the energy consumption optimization information, and send it to the lighting devices and sunshade devices in the building.

[0014] By obtaining the indoor and outdoor lighting images of the building and generating a brightness distribution image, combining with a preset lighting feature positioning method to extract the feature point set of the lighting change area, and then correcting the feature point set based on the distribution information of the stable lighting area, the present invention can effectively improve the accuracy of lighting area recognition and avoid recognition errors caused by local lighting anomalies or environmental interference; further, by performing partition comparison on the lighting areas through the corrected feature point set to obtain the energy consumption optimization information, it provides a more accurate judgment basis for subsequent energy consumption control, thus solving the problems of low image perception accuracy, unstable area recognition, and lagging regulation response in the prior art, realizing the efficient and intelligent management of building lighting and sunshade devices, and improving the overall energy efficiency control level of the building. Brief Description of the Drawings

[0015] Figure 1 It is a flowchart of a building energy consumption control method using lighting image changes provided by the present invention; Figure 2 It is a flowchart of a method for obtaining the energy consumption optimization information of lighting areas provided by the present invention; Figure 3 It is a framework diagram of a building energy consumption control system using lighting image changes provided by the present invention.

[0016] Reference Signs: A building energy consumption control system 100 using lighting image changes, an image processing module 101, a feature positioning module 102, a feature correction module 103, a partition comparison module 104. Detailed Embodiments

[0017] To make the objectives, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below with reference to the accompanying drawings in this application. Apparently, the described embodiments are some, but not all, of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts shall fall within the scope of protection of this application.

[0018] The terms "first" and "second" are used for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features; in the description of this application, unless otherwise stated, the meaning of "plural" is two or more.

[0019] To more clearly illustrate the technical solutions of this invention, the following describes this invention in detail with reference to specific embodiments, but it should not be construed as limiting the scope of protection of this invention.

[0020] In this embodiment, as Figure 1 [[ID=~11]]shown, a building energy consumption control method using the change of illumination images is provided. This method can achieve precise energy consumption control in a complex illumination environment and provide reliable technical support for the energy-saving management of buildings.

[0021] Specifically, step S01: Obtain the indoor and outdoor illumination images of the building and generate a brightness distribution image corresponding to the illumination image. First, collect the illumination images through cameras or illumination sensors installed inside and outside the building. These cameras can be placed in key indoor areas, such as the ceilings of offices or halls, to capture the illumination conditions of indoor natural light and artificial lighting. At the same time, outside the building, cameras can be installed on the exterior wall or roof of the building to record the illumination information of direct sunlight or diffused light on cloudy days. The collected illumination images can be color images or grayscale images, containing the spatial distribution information of indoor and outdoor illumination intensities. Further, to generate the brightness distribution image, process the collected illumination images to extract their brightness information. For example, convert the color image to a grayscale image, with the grayscale value of each pixel representing the illumination intensity, or extract the brightness component from the image to generate an image reflecting the illumination intensity distribution. This brightness distribution image shows the change of illumination in space with the brightness values of pixel points. Areas with high brightness values indicate strong illumination, and areas with low brightness values indicate weak illumination. This processing ensures that subsequent steps can be analyzed based on clear illumination intensity information.

[0022] Further, in step S02: Based on a preset illumination feature localization method, obtain the feature point set of the illumination change region in the brightness distribution image. Analyze the brightness distribution image using the preset illumination feature localization method to identify the illumination change region and extract its feature point set. The illumination change region refers to the region in the brightness distribution image where the illumination intensity changes significantly, such as the boundary region where natural light enters the room through the window and intersects with artificial lighting. Specifically, the preset illumination feature localization method can be implemented by detecting the rapid change of brightness values in the brightness distribution image. For example, analyze the spatial change rate of brightness values to find the regions where the brightness values increase or decrease significantly within a short distance, and these regions usually correspond to the illumination change regions. Then, extract the feature point set within these regions. The feature point set consists of a group of key points, and these points represent the typical positions of the illumination change region, such as the light and shadow boundary or the point with the most significant brightness change. When extracting the feature point set, a set of coordinate points can be generated by identifying the key positions within the brightness change region, and each coordinate point corresponds to a pixel position in the brightness distribution image. These coordinate points can reflect the spatial distribution characteristics of the illumination change region and provide a basis for subsequent analysis.

[0023] Further, in step S03: Obtain the distribution information of the feature point set to the stable illumination region in the brightness distribution image, and correct the feature point set according to the distribution information to obtain the corrected feature point set. Further analyze the relationship between the feature point set and the stable illumination region to improve the accuracy of the feature point set. The stable illumination region refers to the region in the brightness distribution image where the illumination intensity changes less, such as the region far from the window indoors or the region outdoors with uniform illumination. First, identify the stable illumination region in the brightness distribution image, which can be achieved by checking the spatial consistency of brightness values. For example, select the region with less fluctuation of brightness values as the stable illumination region. Next, obtain the distribution information of the feature point set to the stable illumination region, specifically including the spatial position relationship and brightness difference between the feature point set and the stable illumination region. For example, determine the central position of the feature point set and analyze the distance or brightness value difference from the central position to the representative position of the stable illumination region. These distribution information reflect the relative characteristics of the illumination change region and the stable illumination region. Based on these distribution information, correct the feature point set to make it more accurately represent the illumination change region. For example, by adjusting the coordinate positions of the feature point set to make it closer to the brightness characteristics of the stable illumination region, reducing the deviation caused by noise or abnormal points. The corrected feature point set is still a set of coordinate points, but its position can better reflect the true characteristics of the illumination change region.

[0024] Further, in step S04: Based on the corrected feature point set, the lighting areas in the lighting image are partitioned and compared to obtain the energy consumption optimization information for the lighting areas. Specifically, centering on the corrected feature point set, the lighting areas in the lighting image are divided into multiple sub-areas. These sub-areas can be generated based on the spatial distribution of the feature point set by means of uniform division or a method based on feature point density. For example, the lighting area is segmented into several rectangular areas, each area containing some feature points and covering a part of the lighting change area. Then, the brightness distribution of each sub-area is compared to analyze the brightness differences between the sub-areas. The specific method includes checking the distribution characteristics of the brightness values in each sub-area, such as the average level or the degree of change of the brightness values. If the brightness values of some sub-areas are significantly higher or lower than those of other areas, it indicates that the lighting distribution is uneven, and it may be necessary to adjust the lighting equipment or shading devices to optimize the energy consumption. Based on these comparison results, the energy consumption optimization information is generated. The energy consumption optimization information can be a description of the uniformity of the lighting area, such as "the lighting area is uniform" or "the lighting area is uneven", or it can be specific control suggestions, such as "reduce the lighting intensity in a certain area" or "increase the introduction of natural light in a certain area". This information is used to guide the energy consumption management in the building. For example, by reducing the lighting power in the areas with excessive lighting or adjusting the shading devices to optimize the lighting distribution, the energy-saving effect can be achieved.

[0025] It can be understood that the implementation of this method can be executed in real time or periodically. For example, the lighting image is collected once every hour and the above analysis is performed to adapt to the changes in indoor and outdoor lighting conditions. For example, on a sunny day, the natural light may be sufficient, and the generated energy consumption optimization information may suggest reducing the artificial lighting; while on a cloudy day or at night, the lighting change area may be less, and the energy consumption optimization information may suggest increasing the lighting intensity in specific areas. This dynamic adjustment can effectively balance the lighting comfort and energy consumption savings.

[0026] In this embodiment, by acquiring the indoor and outdoor lighting images of the building and generating the brightness distribution images, combining the preset lighting feature localization method to extract the feature point set of the lighting change area, and then correcting the feature point set based on the distribution information of the stable lighting area, the accuracy of lighting area recognition can be effectively improved, and the recognition errors caused by local lighting anomalies or environmental interference can be avoided; further, by performing partition comparison of the lighting area through the corrected feature point set to obtain the energy consumption optimization information, a more accurate judgment basis is provided for subsequent energy consumption control, thus solving the problems of low image perception accuracy, unstable area recognition, and lagging regulation response in the prior art, achieving the efficient and intelligent management of building lighting and shading devices, and improving the overall energy efficiency control level of the building.

[0027] In some embodiments, the step of obtaining the distribution information of the feature point set in the stable illumination region of the luminance distribution image includes: setting a first reference line and a second reference line that intersect at the center of the feature point set, pass through the stable illumination region, and are perpendicular to each other, so that the first reference line intersects the illumination change region at a first intersection point and a second intersection point, and the second reference line intersects the illumination change region at a third intersection point and a fourth intersection point; taking the distances and luminance differences between the center of the feature point set and the first intersection point, the second intersection point, the third intersection point, and the fourth intersection point as the distribution information.

[0028] Specifically, in the step of setting a first reference line and a second reference line that intersect at the center of the feature point set, pass through the stable illumination region, and are perpendicular to each other, so that the first reference line intersects the illumination change region at a first intersection point and a second intersection point, and the second reference line intersects the illumination change region at a third intersection point and a fourth intersection point, first determine the center position of the feature point set. The feature point set consists of a set of coordinate points representing key positions in the illumination change region, such as {(x1, y1), (x2, y2), ..., (x n , y n )}. By calculating the average value of these coordinate points, the center position C(x c , y c ) of the feature point set is obtained. This center position is used as a reference for subsequent setting of the reference lines. Further, identify the stable illumination region in the luminance distribution image. The stable illumination region is a region where the illumination intensity changes less, such as a uniform illumination region far from the window indoors. Based on the center of the feature point set, set two perpendicular reference lines: the first reference line and the second reference line. These two reference lines intersect at the center of the feature point set and ensure that they pass through the stable illumination region. For example, the first reference line can be a horizontal line that extends along the horizontal axis of the luminance distribution image, passes through the center of the feature point set and enters the stable illumination region; the second reference line is a vertical line that extends along the vertical axis and also passes through the center of the feature point set and enters the stable illumination region. Then, determine the intersection points of the first reference line and the illumination change region. The illumination change region is a region where the luminance value changes significantly, such as the region where natural light and artificial lighting meet. When the first reference line passes through the illumination change region, it will intersect the boundary of this region, generating two intersection points, which are respectively called the first intersection point P1(x1, y1) and the second intersection point P2(x2, y2). Similarly, when the second reference line passes through the illumination change region, it will also intersect the boundary of this region, generating a third intersection point P3(x3,y3) and a fourth intersection point P4(x4, y4). The determination of these intersection points can be achieved by analyzing the change of the luminance value along the reference line. For example, when the luminance value suddenly changes from the uniform value in the stable illumination region to the non-uniform value in the illumination change region, it is marked as the intersection point position.

[0029] Further, in the step of taking the distances and brightness differences between the center of the feature point set and the first intersection point, the second intersection point, the third intersection point, and the fourth intersection point as distribution information, the spatial distances and brightness differences between the center of the feature point set and each intersection point are calculated to form the distribution information. Specifically, for the first intersection point P1, the distance from the center C of the feature point set to P1 is calculated. For example, this distance value is determined by the straight-line distance between two points. At the same time, the brightness value of the pixel where the center C of the feature point set is located and the brightness value of the pixel where the first intersection point P1 is located are obtained, and the brightness difference between the two is calculated, that is, the absolute difference of the brightness values. Similarly, for the second intersection point P2, the distance from the center C of the feature point set to P2 and the brightness difference between C and P2 are calculated. For the third intersection point P3 and the fourth intersection point P4, the same calculation process is repeated to obtain the distance and brightness difference from C to P3, and the distance and brightness difference from C to P4, respectively. These distances and brightness differences together constitute the distribution information, reflecting the positional relationship of the center of the feature point set relative to the boundary of the illumination change region and the difference in illumination intensity. For example, the distribution information can be represented as four sets of data, each set of data containing the distance and brightness difference between an intersection point and the center of the feature point set. These information describe the relative position characteristics of the center of the feature point set between the illumination change region and the stable illumination region, providing an accurate reference for subsequent processing.

[0030] It can be understood that the implementation of this step can be dynamically executed during the analysis of the brightness distribution image. For example, when processing an indoor illumination image, the center of the feature point set may be located in the illumination change region near the window, and the stable illumination region may be located deep inside the room. The settings of the first reference line and the second reference line can be adjusted according to the actual illumination distribution of the image to ensure that they can effectively cross the boundary between the stable illumination region and the illumination change region. The determination of the intersection points can also be combined with the change threshold of the brightness value. For example, when the brightness value changes exceed a certain preset value, it is considered that the reference line enters the illumination change region, and thus the positions of the intersection points are marked. The calculation of the distances and brightness differences is completed by directly comparing the pixel coordinates and brightness values, ensuring the accuracy of the results.

[0031] In some embodiments, the step of correcting the feature point set according to the distribution information to obtain a corrected feature point set includes: translating the center of the feature point set along the direction of the first reference line until the first distance and brightness difference between the center of the feature point set and the first intersection point and the second distance and brightness difference between the center of the feature point set and the second intersection point are equal to determine an intermediate feature point set; translating the intermediate feature point set along the direction of the second reference line until the third distance and brightness difference between the intermediate feature point set and the third intersection point and the fourth distance and brightness difference between the intermediate feature point set and the fourth intersection point are equal to determine the corrected feature point set.

[0032] Specifically, in the step of translating the center of the feature point set along the direction of the first reference line until the first distance and brightness difference between the center of the feature point set and the first intersection point and the second distance and brightness difference between the center of the feature point set and the second intersection point are equal to determine the intermediate feature point set, the process is first based on the initial position of the center of the feature point set. The center of the feature point set refers to the average position of a set of coordinate points representing the key positions of the illumination change region, such as C(x c , y c ). The first reference line is a straight line that takes the center of the feature point set as the intersection point and passes through the stable illumination region, such as a horizontal line. The first intersection point P1(x1, y1) and the second intersection point P2(x2, y2) are the two intersection points of the first reference line and the boundary of the illumination change region. The illumination change region is a region where the brightness value changes significantly, such as the region where natural light and artificial lighting meet. Further, calculate the first distance from the center C of the feature point set to the first intersection point P1, that is, the straight-line distance between C and P1, and the brightness difference between C and P1, that is, the difference between the brightness value of the pixel where C is located and the brightness value of the pixel where P1 is located. Similarly, calculate the second distance and brightness difference from C to the second intersection point P2. Then, translate the position of the center C of the feature point set along the direction of the first reference line, such as moving left or right along the horizontal direction. The goal of the translation is to make the first distance and brightness difference between the adjusted center C' of the feature point set and the first intersection point P1 equal to the second distance and brightness difference between C' and the second intersection point P2. Specifically, adjust the position of C so that the distances from C' to P1 and P2 are equal (for example, C' is located at the middle position between P1 and P2), and the brightness difference between the pixel where C' is located and P1 and P2 also reaches balance (for example, the absolute values of the brightness differences are equal or close). After completing this translation, a new center C' of the feature point set is obtained, which is called the center of the intermediate feature point set, and the corresponding feature point set is the intermediate feature point set. The intermediate feature point set is a new set of coordinate points obtained by moving each coordinate point of the original feature point set correspondingly with the translation of the center C.

[0033] Further, in the step of translating the intermediate feature point set along the direction of the second reference line until the third distance and brightness difference between the intermediate feature point set and the third intersection point and the fourth distance and brightness difference between the intermediate feature point set and the fourth intersection point are equal to determine the corrected feature point set, continue to process based on the center C' of the intermediate feature point set. The second reference line is a straight line perpendicular to the first reference line, intersecting at the center of the feature point set and passing through the stable illumination area, such as a vertical line. The third intersection point P3(x3, y3) and the fourth intersection point P4(x4, y4) are two intersection points of the second reference line and the boundary of the illumination change area. Similarly, calculate the third distance from the center C' of the intermediate feature point set to the third intersection point P3, that is, the straight-line distance between C' and P3, and the brightness difference between C' and P3, that is, the difference between the brightness value of the pixel where C' is located and the brightness value of the pixel where P3 is located. Similarly, calculate the fourth distance and brightness difference from C' to the fourth intersection point P4. Then, translate the position of the center C' of the intermediate feature point set along the direction of the second reference line, such as moving up or down along the vertical direction. The goal of the translation is to make the third distance and brightness difference between the adjusted feature point set center C'' and the third intersection point P3 equal to the fourth distance and brightness difference between C'' and the fourth intersection point P4. Specifically, adjust the position of C' so that the distances from C'' to P3 and P4 are equal (for example, C'' is located at the middle position between P3 and P4), and the brightness difference between the pixel where C'' is located and P3 and P4 also reaches balance (for example, the absolute values of the brightness differences are equal or close). After completing this translation, the final adjusted feature point set center C'' is obtained, and the corresponding feature point set is the corrected feature point set. The corrected feature point set is a new set of coordinate points obtained by moving each coordinate point of the intermediate feature point set accordingly with the translation of the center C'.

[0034] It can be understood that during the implementation of this step, the translation process can be carried out step by step to ensure that the adjustment of the feature point set center can balance the distance and brightness difference simultaneously. For example, when translating along the first reference line, by gradually moving the center of the feature point set and comparing the distance and brightness difference with the first intersection point and the second intersection point in real time, a position that satisfies the equal distance and brightness difference can be found. Similarly, when translating along the second reference line, repeat a similar process to ensure that the center of the intermediate feature point set is adjusted to a position where the distance and brightness difference with the third intersection point and the fourth intersection point are equal. This step-by-step adjustment method can improve the accuracy of the corrected feature point set and make it better reflect the characteristics of the illumination change area. In actual operation, the translation step size can be set according to the resolution of the brightness distribution image or the size of the illumination change area, for example, fine-tuned in pixels.

[0035] In some embodiments, the steps of obtaining indoor and outdoor lighting images of a building and generating a brightness distribution image corresponding to the lighting image include: collecting lighting images of the indoor and outdoor environments of the building and performing brightness normalization processing on the lighting images; performing edge enhancement processing on the lighting images after brightness normalization processing to obtain the brightness distribution image.

[0036] Specifically, in the steps of collecting lighting images of the indoor and outdoor environments of the building and performing brightness normalization processing on the lighting images, first, lighting images are obtained by lighting acquisition devices deployed inside and outside the building. These devices can be high-resolution digital cameras or lighting sensors, installed at key positions of the building to capture indoor and outdoor lighting conditions. For example, indoors, the camera can be placed on the ceiling of the main activity areas, such as offices, corridors or meeting rooms, to record the combined lighting effect of natural light and artificial lighting; outdoors, the camera can be installed on the exterior wall of the building, the roof or in the direction facing the main light source, to capture the characteristics of direct sunlight, diffused light or overcast lighting. The collected lighting images can be color images or grayscale images, containing the spatial distribution information of the lighting intensity. Further, brightness normalization processing is performed on the collected lighting images to eliminate the problem of inconsistent brightness value ranges caused by different lighting conditions or device differences. Specifically, brightness normalization processing maps the brightness values of the lighting images to a standardized range, for example, adjusting the brightness value of the pixel to between 0 and 1, or between 0 and 255. The implementation method can be dividing the brightness value of each pixel by the maximum brightness value in the image, or scaling the brightness value to a preset range through linear transformation. This processing ensures that images collected at different times or by different devices have comparable brightness values, facilitating subsequent analysis. For example, images collected during strong sunlight during the day may have higher brightness values, while images collected on cloudy days or at night have lower brightness values, and normalization processing enables the brightness values of these images to be compared on a unified scale.

[0037] Further, in the step of performing edge enhancement on the illumination image after brightness normalization to obtain the brightness distribution image, the normalized illumination image is further processed to highlight the spatial variation characteristics of the illumination intensity, thereby generating the brightness distribution image. Specifically, edge enhancement aims to enhance regions with significant brightness changes in the image, such as the junction between natural light and artificial lighting or the light and shadow transition regions. These regions usually correspond to rapid changes in illumination intensity and are of great significance for subsequent analysis of illumination distribution. Edge enhancement can be achieved by applying image processing techniques, such as using edge detection algorithms or filtering methods to highlight the boundaries of brightness changes. One implementation is to apply a high-pass filter to the normalized illumination image to amplify the rapidly changing parts of the brightness values in space, making the regions with illumination intensity changes more obvious in the image. Another way is to use the gradient calculation method to identify the rate of change of brightness values in the horizontal and vertical directions and highlight the edge regions of brightness changes. After edge enhancement, the obtained image is the brightness distribution image, which clearly shows the spatial distribution characteristics of the illumination intensity with the brightness values of pixel points. In the brightness distribution image, regions with higher brightness values represent areas with stronger illumination, regions with lower brightness values represent areas with weaker illumination, and the enhanced edge regions highlight the boundaries of illumination changes, providing clearer illumination distribution information for subsequent analysis.

[0038] It can be understood that during the implementation of this step, the frequency of collecting the illumination image can be set according to actual needs. For example, it can be collected once an hour to adapt to daylight changes, or collected when a specific event is triggered (such as a significant change in illumination intensity). The brightness normalization process can be adjusted according to the resolution of the collection device or the dynamic range of the illumination environment. For example, in a high-dynamic-range environment, more complex normalization methods may be required to retain details. The intensity of edge enhancement can also be optimized according to the characteristics of the illumination image. For example, in an indoor environment with relatively smooth illumination changes, weaker edge enhancement can be used to avoid over-amplifying noise; while in an outdoor environment with strong illumination contrast, stronger edge enhancement can be used to highlight the light and shadow boundaries. These adjustments ensure that the brightness distribution image can accurately reflect the illumination distribution characteristics inside and outside the building.

[0039] In some embodiments, the step of obtaining the feature point set of the illumination change region in the brightness distribution image based on the preset illumination feature positioning method includes: analyzing the brightness distribution image based on the preset illumination feature positioning method; when the feature point set of the illumination change region in the brightness distribution image is recognized, the feature point set is used as the initial feature point set, and the steps are executed: obtaining the distribution information of the feature point set to the stable illumination region in the brightness distribution image; when the feature point set of the illumination change region in the brightness distribution image cannot be recognized, the corresponding prompt information is output.

[0040] Specifically, in the step of analyzing the brightness distribution image based on a preset illumination feature location method, the preset illumination feature location method is used to process the brightness distribution image to identify illumination variation regions and extract their feature point sets. A brightness distribution image is an image that reflects the spatial distribution of light intensity, where the brightness value of a pixel represents the light intensity. Illumination variation regions are areas in the brightness distribution image where light intensity changes significantly, such as the boundary where natural light entering a room through a window meets artificial lighting. The preset illumination feature location method is an image analysis-based technique designed to detect significant changes in brightness values in a brightness distribution image. Specifically, this method analyzes the variation characteristics of brightness values in image space, for example, by examining the differences in brightness values between adjacent pixels to identify regions of rapid brightness changes. These regions typically correspond to transitional areas of light intensity, such as light-shadow boundaries or areas where light intensity gradually changes from high to low. During the analysis process, each pixel in the brightness distribution image can be scanned and its brightness value compared with the brightness values of surrounding pixels. If the difference exceeds a preset threshold, the region is marked as an illumination variation region. This analysis method can effectively distinguish areas with changing lighting from areas with relatively uniform lighting intensity, providing a basis for the subsequent extraction of feature point sets.

[0041] Furthermore, when a feature point set of an illumination change region in the brightness distribution image is identified, the feature point set is used as an initial feature point set, and the step of obtaining distribution information of the feature point set to a stable illumination region in the brightness distribution image is performed. If the analysis result indicates that an illumination change region exists in the brightness distribution image, feature point sets within these regions are further extracted. A feature point set is a set of coordinate points representing key positions of an illumination change region, such as {(x1, y1), (x2, y2), ..., (x n , y n )}, these points are usually located at locations where brightness changes significantly, such as the boundary of the illumination change area or the point where the brightness value changes most dramatically. When extracting the feature point set, key points can be found in the identified illumination change area, such as selecting the pixel point with the highest rate of change in brightness value as the feature point, or selecting a representative point on the boundary of the illumination change area. After the feature point set is extracted, it is used as the initial feature point set for subsequent processing steps. Specifically, the initial feature point set will be used to obtain its distribution information with the stable illumination area, which refers to the area with small changes in illumination intensity in the brightness distribution image, such as the uniformly illuminated area indoors away from the window. This subsequent step involves analyzing the spatial and brightness relationship between the initial feature point set and the stable illumination area, but this embodiment only focuses on the extraction of the feature point set and its confirmation as the initial feature point set.

[0042] Further, in the step of outputting corresponding prompt information when the feature point set of the illumination change area in the brightness distribution image cannot be recognized, if the analysis result shows that there is no obvious illumination change area in the brightness distribution image or an effective feature point set cannot be extracted, prompt information is generated and output to notify the subsequent processing flow or the user. For example, in some cases, such as in a fully artificial lighting environment with uniform lighting indoors at night or in a cloudy environment with completely uniform outdoor lighting, the brightness distribution image may show relatively consistent brightness values and lack significant illumination change areas. At this time, the preset illumination feature localization method may not be able to detect areas where the brightness value changes exceed the threshold, or the detected change areas are too weak to form an effective feature point set. In this case, prompt information is output, and the prompt information can be a text description, such as "No illumination change area detected" or "Unable to extract feature point set". This prompt information can be used to trigger an alternative processing flow, such as pausing subsequent analysis steps, or suggesting re-acquiring the illumination image to obtain more suitable data. The output method of the prompt information can be recorded in a log file or displayed through a user interface, depending on the implementation environment.

[0043] It can be understood that during the implementation of this step, the threshold of the preset illumination feature localization method can be adjusted according to the characteristics of the brightness distribution image. For example, in an environment with high illumination contrast, a higher brightness change threshold can be set to avoid false detection of noise; in an environment with relatively smooth illumination changes, the threshold can be lowered to capture weak changes. The extraction of the feature point set can also be optimized according to the size and complexity of the illumination change area. For example, more feature points can be extracted in a larger illumination change area to improve representativeness, while fewer feature points can be extracted in a smaller area to reduce the computational amount. The generation and output of the prompt information can be customized according to the actual application scenario. For example, in a real-time monitoring system, the prompt information may need to be immediately feedback, while in periodic analysis, the prompt information can be accumulated and processed uniformly.

[0044] In some embodiments, as Figure 2 shown, the step of partitioning and comparing the illumination area in the illumination image based on the corrected feature point set to obtain the energy consumption optimization information of the illumination area includes: S61: Centering on the corrected feature point set, dividing the illumination area in the illumination image into a preset number of equal partitions; S62: Obtaining the brightness distribution information of the equal partitions, and determining the energy consumption optimization information of the illumination area according to the differences in the brightness distribution information between the equal partitions.

[0045] Specifically, in the step of dividing the illumination region in the illumination image into a preset number of equal partitions centered on the set of corrected feature points, first, the center of the set of corrected feature points is used as the reference point for division. The set of corrected feature points is a set of adjusted coordinate points, such as {(x1', y1'), (x2', y2'),..., (x n ', y n ')}, which represents the optimized positions of the illumination change region, and its center C''(x c'' , y c'' ) is obtained by calculating the average value of all coordinate points. The illumination image is an image reflecting the illumination intensity inside and outside the building, and the illumination region is the region in this image that contains illumination changes, such as the region where natural light and artificial lighting intersect. Based on the center C'' of the set of corrected feature points, the illumination region is divided into a preset number of equal partitions. The equal partitions refer to dividing the illumination region into sub-regions with similar sizes or shapes, and the preset number can be set according to actual needs, such as 4, 8, or 16 equal partitions. The specific division method can be to divide the illumination region with the center C'' as the origin according to a uniform angle or distance. For example, the illumination region can be divided into four quadrants, with each quadrant being an equal partition, located in the upper left, upper right, lower left, and lower right regions of the center C'' respectively; or with the center C'' as the center of the circle, the illumination region is divided into multiple fan-shaped regions, and each fan-shaped region covers the same angular range. When dividing, ensure that each equal partition contains a part of the illumination region and covers the distribution range of the set of corrected feature points as much as possible, so that the subsequent analysis can comprehensively reflect the characteristics of the illumination region.

[0046] Further, in the step of obtaining the brightness distribution information of the equal partitions and determining the energy consumption optimization information of the illumination area based on the differences in the brightness distribution information between the equal partitions, the brightness distribution of each equal partition is analyzed, and the differences between the equal partitions are compared to generate the energy consumption optimization information. Specifically, the brightness distribution information reflects the spatial characteristics of the illumination intensity within each equal partition, such as the average level or the degree of variation of the brightness values. To obtain the brightness distribution information, the brightness values of the pixels within each equal partition are analyzed, and their statistical characteristics are calculated, such as the average value of all pixel brightness values or the range of brightness values. The brightness values can be directly extracted from the illumination image, such as the pixel values in a grayscale image or the brightness component after conversion of a color image. Then, the brightness distribution information between the equal partitions is compared to identify the brightness differences. For example, if the average brightness of a certain equal partition is significantly higher than that of other partitions, it may indicate that this area is exposed to stronger natural light or the artificial lighting is too strong; conversely, if the average brightness of a certain equal partition is significantly lower than that of other partitions, it may indicate that the illumination in this area is insufficient. The comparison of the differences can be achieved by checking the relative deviation of the average brightness of each equal partition or the difference in the range of brightness values. Based on these differences, the energy consumption optimization information of the illumination area is determined. The energy consumption optimization information is a description or control suggestion regarding the illumination distribution state of the illumination area, such as "the illumination area is uniform, it is recommended to reduce the overall illumination intensity" or "the illumination area is non-uniform, it is recommended to increase the illumination of a certain partition or adjust the sunshade device". These information reflect the energy consumption optimization potential of the illumination area, such as achieving energy conservation by reducing the lighting power in areas with excessive illumination or increasing the illumination intensity in areas with insufficient illumination, or reducing the demand for artificial lighting by adjusting the sunshade device to introduce more natural light.

[0047] It can be understood that during the implementation of this step, the number and division method of the equal partitions can be adjusted according to the size and complexity of the illumination area. For example, in an area with more complex illumination changes, the number of equal partitions can be increased (such as divided into 16 partitions) to improve the analysis accuracy; in an area with simpler illumination changes, the number of partitions can be reduced (such as divided into 4 partitions) to reduce the computational complexity. The acquisition and comparison of the brightness distribution information can be optimized according to specific requirements. For example, only the average brightness can be compared, or a more detailed analysis can be carried out by combining the spatial distribution characteristics of the brightness values. The generation of the energy consumption optimization information can be customized according to the actual application scenario. For example, in an office building, the optimization information may tend to adjust the power of the lighting equipment, while in a residential building, more attention may be paid to the adjustment of the sunshade device. These adjustments ensure the practicality and pertinence of the energy consumption optimization information.

[0048] In some embodiments, the brightness distribution information includes a brightness mean value and a gradient value. The step of obtaining the brightness distribution information of the equal partitions and determining the energy consumption optimization information of the illumination area according to the differences in the brightness distribution information between the equal partitions includes: statistically calculating the brightness mean value and the gradient value of each equal partition, and calculating a brightness difference index between the equal partitions. The brightness difference index is determined by the following formula: where ΔL is the brightness difference index, M i is the brightness mean value of the i-th equal partition, M avg is the average brightness mean value of all equal partitions, G i is the brightness gradient value of the i-th equal partition, G avg is the average brightness gradient value of all equal partitions, D i is the distance between the center of the i-th equal partition and the building reference point, D max is the maximum building reference distance, and N is the number of equal partitions; if the brightness difference index is within a preset range, the energy consumption optimization information is that the illumination area is uniform, and an energy-saving control instruction is generated; if the brightness difference index is not within the preset range, the energy consumption optimization information is that the illumination area is non-uniform, and a dynamic adjustment instruction is generated.

[0049] Specifically, in the step of statistically calculating the brightness mean value and the gradient value of each equal partition and calculating the brightness difference index between the equal partitions, first, the brightness distribution information of each equal partition is analyzed. The equal partitions are centered on the corrected feature point set, and the illumination area is divided into a preset number of sub-regions, such as 4 or 8 regions of similar size. The brightness distribution information includes a brightness mean value and a gradient value, where the brightness mean value reflects the average level of the illumination intensity within the partition, and the gradient value reflects the severity of the brightness change within the partition. For each equal partition, calculate its brightness mean value M i , that is, the average value of all pixel brightness values within the partition. The brightness value can be extracted from the gray value or the brightness component of the illumination image. For example, in a gray image, the gray value of the pixel is directly used. Then, calculate the brightness gradient value G i of each equal partition. The gradient value represents the rate of change of brightness in space and is usually obtained by analyzing the differences in pixel brightness values in the horizontal and vertical directions within the partition. For example, the brightness difference between each pixel and its neighboring pixels can be calculated, and the average or maximum value of these differences can be taken as the gradient value of the partition. After completing the statistics of the brightness mean value and the gradient value of each equal partition, calculate the average brightness mean value M avg of all equal partitions (i.e., the average value of all M i ) and the average brightness gradient value G avg of all equal partitions (i.e., the average value of all G i ). Further, determine the distance D between the center of each equal partition and the building reference pointi , the building reference point can be a fixed position within the building, such as the center of the entrance or the geometric center of the illumination image; D max is the maximum distance from the centers of all possible partitions within the building to the reference point, used to normalize the distance effect. Based on this data, the luminance difference index ΔL is calculated, which comprehensively considers the differences in luminance mean, gradient value, and distance. Specifically, the luminance difference index is determined as follows: for each equal partition, calculate the square of the difference between its luminance mean and the average luminance mean, and the square of the difference between its gradient value and the average gradient value, sum them up and take the average, then multiply by a distance-based weight ( ), and finally take the square root. This index reflects the overall difference degree of the luminance distribution in each equal partition, and the larger the value, the more significant the difference.

[0050] Further, in the step of generating an energy-saving control instruction if the luminance difference index is within a preset range and the energy consumption optimization information is that the illumination area is uniform, check whether the calculated luminance difference index ΔL is within the preset range. The preset range can be set according to the actual application scenario. For example, 0 to 0.1 indicates uniform illumination distribution. If ΔL is within this range, it means that the luminance means and gradient values of each equal partition are relatively close, and the illumination intensity distribution in the illumination area is uniform. At this time, generate energy consumption optimization information indicating that the illumination area is uniform and suitable for taking energy-saving measures. Based on this information, generate an energy-saving control instruction, which is a specific operation suggestion for reducing energy consumption, such as reducing the power of the overall lighting equipment or adjusting the shading device to reduce unnecessary natural light introduction. These instructions aim to minimize energy consumption while maintaining illumination comfort.

[0051] Further, in the step of generating a dynamic adjustment instruction if the luminance difference index is not within the preset range and the energy consumption optimization information is that the illumination area is non-uniform, if the luminance difference index ΔL exceeds the preset range, such as being greater than 0.1, it means that there are significant differences in the luminance means or gradient values of each equal partition, and the illumination intensity distribution in the illumination area is non-uniform. For example, the equal partitions near the window may have a higher luminance mean, while the partitions far from the window have a lower luminance mean. At this time, generate energy consumption optimization information indicating that the illumination area is non-uniform and specific areas need to be adjusted for illumination. Based on this information, generate a dynamic adjustment instruction, which is a local adjustment suggestion for non-uniform illumination distribution, such as increasing the illumination intensity of the under-illuminated partition or adjusting the shading device to balance the natural light distribution. These instructions aim to improve the uniformity of the illumination distribution while optimizing energy consumption.

[0052] Understandably, during the implementation of this step, the calculation of the brightness mean and gradient value can be optimized according to the resolution of the illumination image and the partition size. For example, in a high-resolution image, more pixels within each partition can be sampled to improve the accuracy of the mean and gradient values; while in a low-resolution image, the number of sampling points can be reduced to lower the computational complexity. The preset range of the brightness difference index can be adjusted according to the building type and lighting requirements. For example, in an office environment, a more stringent uniformity range may be required, while in a warehouse environment, it can be appropriately relaxed. The generation of the energy-saving control instruction and the dynamic adjustment instruction can be combined with the actual device status. For example, lighting devices with lower power consumption or more easily operable sunshade devices can be preferentially selected to improve the execution efficiency.

[0053] In some embodiments, the step of generating the energy-saving control instruction or the dynamic adjustment instruction includes: when the energy consumption optimization information indicates that the illumination area is uniform, generating an energy-saving control instruction to reduce the power of the lighting device or adjust the opening / closing angle of the sunshade device; when the energy consumption optimization information indicates that the illumination area is non-uniform, generating a dynamic adjustment instruction to increase the brightness of the lighting device or adjust the angle of the sunshade device for a specific partition.

[0054] Specifically, in the step of generating an energy-saving control instruction to reduce the power of the lighting device or adjust the opening / closing angle of the sunshade device when the energy consumption optimization information indicates that the illumination area is uniform, first, it is processed according to the judgment result of the energy consumption optimization information. The energy consumption optimization information indicates that the illumination area is uniform, that is, the illumination intensity distribution in each equal partition is relatively consistent. For example, the difference in the brightness mean and gradient value between each partition is small. In this case, the overall illumination intensity of the illumination area may be high or sufficient to meet the usage requirements. Therefore, the energy use can be optimized by reducing the energy consumption. Based on this, an energy-saving control instruction is generated, which specifically includes two operation suggestions: one is to reduce the power of the lighting device, and the other is to adjust the opening / closing angle of the sunshade device. For reducing the power of the lighting device, the instruction can be targeted at the lighting system in the building, such as LED lights or fluorescent lights, and it is recommended to reduce the power to a certain level. For example, adjust the current power from 100% to 70% to reduce the power consumption of electricity while maintaining sufficient lighting brightness. For adjusting the opening / closing angle of the sunshade device, the instruction can be targeted at curtains, blinds or other sunshade devices, and it is recommended to reduce the opening / closing angle to reduce the entry of natural light. For example, adjust the blind angle from fully open to half open to avoid excessive strong light. This adjustment is especially applicable to the scenario where natural light is sufficient during the day. By reducing artificial lighting or restricting the entry of excessive natural light, the energy-saving goal can be achieved. The specific content of the energy-saving control instruction can be customized according to the actual lighting environment and device status of the building. For example, lighting devices with larger power consumption or easily operable sunshade devices can be preferentially selected.

[0055] Furthermore, in the step of generating a dynamic adjustment instruction to increase the brightness of lighting equipment or adjust the angle of a sunshade device for a specific partition when the energy consumption optimization information indicates uneven lighting areas, it is processed based on the situation where the energy consumption optimization information shows uneven lighting areas. Uneven lighting areas mean that the lighting intensity in some equal partitions is significantly lower or higher than that in other partitions. For example, the partitions near the window have stronger lighting, while the partitions far from the window have weaker lighting. In response to this situation, dynamic adjustment instructions are generated, aiming to improve the uniformity of lighting distribution through local adjustment while optimizing energy consumption. The dynamic adjustment instructions include two operation suggestions: one is to increase the brightness of lighting equipment for a specific partition, and the other is to adjust the angle of the sunshade device. For increasing the brightness of lighting equipment, the instruction can specify a specific partition with insufficient lighting, such as the corner area far from the window indoors, and suggest increasing the brightness of the lighting equipment in this area. For example, increase the brightness of the LED lights from 50% to 80% to ensure that the lighting intensity in this partition reaches a comfortable level. For adjusting the angle of the sunshade device, the instruction can be for partitions with insufficient or excessive lighting, and suggest adjusting the opening and closing angle of the curtain or blind. For example, in partitions with insufficient lighting, increase the opening and closing angle of the sunshade device to introduce more natural light, such as adjusting the blind from half-open to fully open; in partitions with excessive lighting, reduce the opening and closing angle to reduce glare, such as adjusting the curtain from fully open to partially blocking. These dynamic adjustment instructions target the lighting characteristics of specific partitions, ensuring that the adjustment measures are accurate and effective while avoiding unnecessary energy consumption increase.

[0056] Understandably, during the implementation of this step, the generation of energy-saving control instructions and dynamic adjustment instructions can be dynamically adjusted according to the actual equipment configuration and lighting environment of the building. For example, in an office building, the energy-saving control instructions may prioritize reducing the lighting power in meeting rooms or lobbies, while the dynamic adjustment instructions may target specific workstation areas to adjust the lighting brightness. In a residential environment, the instructions may focus more on adjusting the sunshade device to balance the utilization of natural light. The execution of the instructions can be combined with the intelligent control system of the building. For example, the adjustment can be automatically achieved through controllers connected to lighting equipment and sunshade devices, or manually executed by prompting operators through a user interface. In addition, the generation of the instructions can consider time factors. For example, during the day, the sunshade device is preferentially adjusted to utilize natural light, while at night, the brightness of lighting equipment is mainly adjusted. These customized measures ensure the practicality and efficiency of the instructions.

[0057] The present invention also provides another embodiment, as Figure 3As shown in the figure, a building energy consumption control system 100 that utilizes changes in illumination images. Specifically, the system 100 includes: an image processing module 101, which is used to obtain indoor and outdoor illumination images of the building and generate a brightness distribution image corresponding to the illumination image; a feature localization module 102, which is used to obtain a set of feature points in the illumination change region in the brightness distribution image based on a preset illumination feature localization method; a feature correction module 103, which is used to obtain the distribution information of the set of feature points to the stable illumination region in the brightness distribution image and correct the set of feature points according to the distribution information to obtain a corrected set of feature points; and a partition comparison module 104, which is used to perform partition comparison on the illumination regions in the illumination image based on the corrected set of feature points to obtain energy consumption optimization information for the illumination regions.

[0058] Further, for the image processing module 101, its function is to obtain indoor and outdoor illumination images of the building and generate a brightness distribution image corresponding to the illumination image. First, illumination images of the indoor and outdoor environments are collected through illumination collection devices deployed inside and outside the building, such as high-resolution digital cameras or illumination sensors. The indoor camera can be installed on the ceiling of the main activity areas, such as offices, meeting rooms or halls, to capture the combined illumination effect of natural light and artificial lighting; the outdoor camera can be installed on the exterior wall, roof of the building or at a position facing the main light source to record the characteristics of direct sunlight, diffused light or cloudy day illumination. The collected illumination images can be color images or grayscale images, containing the spatial distribution information of the illumination intensity. Further, the image processing module 101 processes the collected illumination images to generate a brightness distribution image. Specifically, the image processing module 101 extracts the brightness information of the image, for example, by converting a color image into a grayscale image and using the grayscale value of each pixel to represent the illumination intensity, or by extracting the brightness component from the color image. The processed result is a brightness distribution image, where the brightness value of the pixel reflects the spatial distribution of the illumination intensity. Regions with high brightness values indicate stronger illumination, and regions with low brightness values indicate weaker illumination. The image processing module 101 ensures that the generated brightness distribution image can clearly display the illumination characteristics inside and outside the building, providing an accurate data basis for subsequent modules.

[0059] Furthermore, for the feature localization module 102, its function is to obtain a set of feature points in the illumination change region in the brightness distribution image based on a preset illumination feature localization method. The illumination change region refers to the region in the brightness distribution image where the illumination intensity changes significantly, such as the region where natural light enters the room through a window and meets artificial lighting. The feature localization module 102 analyzes the brightness distribution image through the preset illumination feature localization method to identify these illumination change regions. Specifically, this module 102 scans the pixels of the brightness distribution image and detects the spatial change of the brightness value. For example, by comparing the brightness value differences of adjacent pixels, it finds the regions where the brightness changes rapidly. These regions usually correspond to the light and shadow boundaries or the illumination intensity transition regions. Then, the feature localization module 102 extracts a set of feature points within the illumination change region. The set of feature points is a group of coordinate points representing the key positions of the illumination change region, such as {(x1, y1), (x2, y2),..., (x n , y n )}. These feature points are usually located at the positions where the brightness changes significantly, such as the boundaries of the illumination change region or the points where the brightness value changes most violently. When extracting the feature points, the feature localization module 102 can select the pixel points with the highest brightness change rate or the representative points on the region boundary as the feature points. The processing result of the feature localization module 102 is a set of coordinate points, which accurately reflects the spatial characteristics of the illumination change region and provides a basis for the analysis of subsequent modules.

[0060] Furthermore, for the feature correction module 103, its function is to obtain the distribution information of the set of feature points to the stable illumination region in the brightness distribution image, and correct the set of feature points according to this distribution information to obtain a corrected set of feature points. The stable illumination region refers to the region in the brightness distribution image where the illumination intensity changes less, such as the uniform illumination region far from the window indoors or the region with uniform illumination outdoors. The feature correction module 103 first identifies the stable illumination region in the brightness distribution image. For example, by analyzing the spatial consistency of the brightness value, it selects the region with less fluctuation of the brightness value. Then, the feature correction module 103 calculates the distribution information between the set of feature points and the stable illumination region, including the spatial distance and brightness difference from the center of the set of feature points to the stable illumination region. For example, the feature correction module 103 determines the center position C(x c , y c ) of the set of feature points, and calculates the distance from it to the representative position of the stable illumination region, as well as the brightness value difference between the pixel at the center of the set of feature points and the pixels in the stable region. Based on these distribution information, the feature correction module 103 corrects the set of feature points, adjusts the feature point coordinates to reduce the influence of noise or abnormal points. For example, by making the coordinates of the set of feature points approach the brightness characteristics of the stable illumination region, it generates a corrected set of feature points {(x1', y1'), (x2', y2'),..., (x n ', yn ')}. The corrected feature point set more accurately reflects the characteristics of the light change area, providing optimized data for subsequent zonal comparison.

[0061] Furthermore, for the zonal comparison module 104, its function is to perform zonal comparison on the light areas in the light image based on the corrected feature point set to obtain energy consumption optimization information for the light areas. The light area is the area in the light image that contains light changes, such as the area where natural light and artificial lighting meet. The zonal comparison module 104 divides the light area into multiple sub-areas with the center of the corrected feature point set as the reference, for example, generating several rectangular or fan-shaped areas through uniform division. Then, the zonal comparison module 104 analyzes the brightness distribution characteristics of each sub-area, such as calculating the average level or variation degree of the pixel brightness values within each sub-area, and compares the brightness differences between these sub-areas. If the brightness values of some sub-areas are significantly higher or lower than those of other sub-areas, it indicates that the light distribution is uneven and the lighting or shading devices may need to be adjusted. Based on the comparison results, the zonal comparison module 104 generates energy consumption optimization information, which describes the uniformity state of the light area or provides specific energy consumption optimization suggestions, such as "The light area is uniform, it is recommended to reduce the lighting power" or "The light area is uneven, it is recommended to enhance the lighting in a certain area". This information is used to guide the energy consumption management in the building, such as optimizing the power of lighting devices or adjusting the shading devices to balance the light distribution.

[0062] It can be understood that during the implementation of the system 100, each module can work together to achieve real-time or periodic energy consumption control. For example, the image processing module 101 can collect a light image once an hour, and then the feature location module 102 and the feature correction module 103 process the data, and the zonal comparison module 104 generates energy consumption optimization information. The processing parameters of each module can be adjusted according to the building environment. For example, in a commercial building with frequent light changes, the collection frequency can be increased, or in a residential environment with stable light, the processing complexity can be reduced. The system 100 can be implemented through hardware devices (such as cameras, processors) and software algorithms (such as image processing programs) to ensure efficient operation.

[0063] Optionally, the zonal comparison module may further include: a zoning unit for dividing the light area in the light image into a preset number of equal zones with the corrected feature point set as the center; an analysis unit for obtaining the brightness distribution information of the equal zones and determining the energy consumption optimization information for the light area according to the differences in the brightness distribution information between the equal zones; and an instruction generation unit for generating an energy-saving control instruction or a dynamic adjustment instruction according to the energy consumption optimization information and sending it to the lighting devices and shading devices in the building.

[0064] Specifically, for the partitioning unit, its function is to divide the illuminated area in the illumination image into a preset number of equal partitions centered around the corrected feature point set. The corrected feature point set is a set of optimized coordinate points, such as {(x1', y1'), (x2', y2'),..., (x n ', y n ')}, which represents the key positions in the illumination change area. Its center C''(x c'' , y c'' ) is obtained by calculating the average value of all coordinate points. The illumination image is an image that reflects the illumination intensity inside and outside the building, and the illuminated area is the area that contains illumination changes, such as the area where natural light and artificial lighting intersect. The partitioning unit divides the illuminated area into a preset number of equal partitions based on the center C'' of the corrected feature point set. The equal partitions refer to sub-regions with similar sizes or shapes. The preset number can be set according to actual needs, such as dividing into 4, 8, or 16 equal partitions. The specific partitioning method can be to take the center C'' as the origin and divide the illuminated area according to a uniform angle or distance. For example, the illuminated area can be divided into four quadrants, with each quadrant serving as an equal partition, located in the upper left, upper right, lower left, and lower right areas of the center C'' respectively; or with C'' as the center, divide the illuminated area into multiple fan-shaped areas, each fan covering the same angular range. When partitioning, ensure that each equal partition contains a part of the illuminated area and covers the distribution range of the corrected feature point set as much as possible, so that the subsequent analysis can comprehensively reflect the characteristics of the illuminated area. The processing result of the partitioning unit is a set of partitioned sub-regions, providing a data basis for the analysis unit.

[0065] Furthermore, for the analysis unit, its function is to obtain the brightness distribution information of the equally divided regions and determine the energy consumption optimization information of the illumination area based on the differences in the brightness distribution information between the equally divided regions. The brightness distribution information reflects the spatial characteristics of the illumination intensity within each equally divided region, such as the average level or degree of variation of the brightness values. The analysis unit processes each equally divided region to extract its brightness distribution information. Specifically, for each equally divided region, the analysis unit scans the brightness values of the pixels within the region and calculates the statistical characteristics of the brightness values, such as the average value of all pixel brightness values or the range of brightness values. The brightness values can be directly extracted from the illumination image, such as using the pixel values of a grayscale image or the brightness component after converting a color image. Then, the analysis unit compares the brightness distribution information between the equally divided regions to identify the differences. For example, if the average brightness of a certain equally divided region is significantly higher than that of other regions, it may indicate that this region is exposed to strong natural light or has excessive artificial lighting; if the average brightness of a certain equally divided region is significantly lower than that of other regions, it may indicate insufficient illumination. The comparison of the differences can be achieved by checking the relative deviation of the average brightness of each equally divided region or the differences in the range of brightness values. Based on these differences, the analysis unit generates energy consumption optimization information, which describes the uniformity state of the illumination area or provides specific energy consumption optimization suggestions. For example, if the brightness distributions of the equally divided regions are close, the generated information can be "The illumination area is uniform, suitable for reducing the overall energy consumption"; if the brightness of some regions is significantly higher or lower, the generated information can be "The illumination area is uneven, and the illumination needs to be adjusted for specific regions". The energy consumption optimization information provides a decision-making basis for the instruction generation unit.

[0066] Furthermore, for the instruction generation unit, its function is to generate energy-saving control instructions or dynamic adjustment instructions based on the energy consumption optimization information and send them to the lighting equipment and shading devices in the building. If the energy consumption optimization information indicates that the lighting area is uniform, the instruction generation unit generates energy-saving control instructions and suggests operations to reduce energy consumption. For example, the instruction can suggest reducing the power of the lighting equipment, such as adjusting the power of the LED lamp from 100% to 70% to reduce power consumption; or suggest adjusting the opening and closing angle of the shading device, such as adjusting the louvers from fully open to half open to reduce the entry of excessive natural light. These energy-saving control instructions are applicable to scenarios with sufficient and uniform lighting, aiming to reduce energy consumption while maintaining lighting comfort. If the energy consumption optimization information indicates that the lighting area is uneven, the instruction generation unit generates dynamic adjustment instructions to adjust the lighting problems in specific partitions. For example, the instruction can suggest increasing the brightness of the lighting equipment in the underlit partition, such as increasing the brightness of the lamps in a certain area from 50% to 80%; or adjusting the angle of the shading device, such as adjusting the curtain from half open to fully open in the underlit partition to introduce more natural light, or partially closing the curtain in the overlit partition to reduce glare. The instruction generation unit sends these instructions to the lighting equipment and shading devices in the building through the communication interface, such as connecting to the lamp controller or shading motor through the intelligent control system to achieve automatic adjustment, or prompting the operator to execute manually through the user interface.

[0067] It can be understood that in the implementation process of the partition comparison module, each unit can work together to achieve efficient energy consumption optimization. For example, the partition unit can dynamically adjust the number of equal partitions according to the complexity of the lighting area, such as increasing the number of partitions in areas with large lighting changes to improve the analysis accuracy. The analysis unit can optimize the extraction method of brightness distribution information, such as improving the accuracy by sampling more pixels, or reducing the sampling points in low-resolution images to reduce the calculation amount. The instructions generated by the instruction generation unit can be customized according to the building equipment configuration, such as preferentially selecting dimmable LED lights for power adjustment, or preferentially adjusting the electric shading device to simplify the operation. The instructions can be sent in real time or in batches at specific time intervals (such as every hour) to adapt to the changes in the lighting environment.

[0068] The above description is only an exemplary implementation manner of the present invention, and does not limit the patent scope of the present invention. Any equivalent structural transformation made under the technical concept of the present invention by using the content of the specification and drawings of the present invention, or directly / indirectly applied to other related technical fields, is included in the patent protection scope of the present invention.

Claims

1. A building energy consumption control method using changes in illumination patterns, characterized in that: The building energy consumption control method comprises the following steps: Acquire indoor and outdoor lighting images of the building, and generate a brightness distribution image corresponding to the lighting image; Based on a preset illumination feature positioning method, a feature point set of the illumination change area in the brightness distribution image is obtained; Acquiring distribution information of the feature point set to a stable illumination area in the brightness distribution image, and correcting the feature point set according to the distribution information to obtain a corrected feature point set; Based on the corrected feature point set, the illumination area in the illumination image is partitioned and compared to obtain energy consumption optimization information of the illumination area.

2. The building energy consumption control method using illumination pattern changes according to claim 1, characterized in that: The step of obtaining distribution information of the feature point set to the stable illumination area in the brightness distribution image includes: Setting a first reference line and a second reference line having the center of the feature point set as the intersection point, passing through the stable illumination area and being perpendicular to each other, so that the first reference line intersects the illumination variation area at the first and second intersection points, and the second reference line intersects the illumination variation area at the third and fourth intersection points; The distances and brightness differences between the center of the feature point set and the first intersection point, the second intersection point, the third intersection point, and the fourth intersection point are used as distribution information.

3. The building energy consumption control method using illumination pattern changes according to claim 2, characterized in that: The step of correcting the feature point set according to the distribution information to obtain a corrected feature point set includes: translating the center of the feature point set along the direction of the first reference line until a first distance and a brightness difference between the center of the feature point set and the first intersection point and a second distance and a brightness difference between the center of the feature point set and the second intersection point are equal, to determine an intermediate feature point set; The intermediate feature point set is translated along the direction of the second reference line until a third distance and a brightness difference between the intermediate feature point set and the third intersection point are equal to a fourth distance and a brightness difference between the intermediate feature point set and the fourth intersection point, so as to determine a revised feature point set.

4. The building energy consumption control method using illumination pattern changes according to claim 1, characterized in that: The step of acquiring indoor and outdoor lighting images of a building and generating a brightness distribution image corresponding to the lighting image comprises: Collecting illumination images of the indoor and outdoor environments of a building, and performing brightness normalization processing on the illumination images; The illumination image after brightness normalization is subjected to edge enhancement processing to obtain the brightness distribution image.

5. The building energy consumption control method using illumination pattern changes according to claim 1, characterized in that: The step of obtaining a feature point set of the illumination change area in the brightness distribution image based on a preset illumination feature positioning method includes: Analyzing the brightness distribution image based on a preset illumination feature positioning method; When a feature point set of an illumination change region in the brightness distribution image is identified, the feature point set is used as an initial feature point set, and the steps of: obtaining distribution information of the feature point set to a stable illumination region in the brightness distribution image; When the feature point set of the illumination change area in the brightness distribution image cannot be identified, corresponding prompt information is output.

6. The building energy consumption control method using illumination pattern changes according to any one of claims 1 to 5, characterized in that: The step of performing partition comparison on the illumination area in the illumination image based on the modified feature point set to obtain energy consumption optimization information of the illumination area includes: Dividing the illuminated area in the illuminated image into a preset number of equal partitions with the corrected feature point set as the center; Brightness distribution information of the equal subareas is obtained, and energy consumption optimization information of the illumination area is determined according to differences in the brightness distribution information between the equal subareas.

7. The building energy consumption control method using illumination pattern changes according to claim 6, characterized in that: The brightness distribution information includes a brightness mean and a gradient value. The step of obtaining the brightness distribution information of the equal partitions and determining the energy consumption optimization information of the illuminated area based on the difference in the brightness distribution information between the equal partitions includes: counting the brightness mean and gradient value of each equal partition, and calculating a brightness difference index between the equal partitions, wherein the brightness difference index is determined by the following formula: Among them, ΔL is the brightness difference index, M i is the mean brightness of the ith equal partition, M avg is the average brightness of all equal partitions, G i is the brightness gradient value of the i-th equal partition, G avg is the average brightness gradient value of all equal partitions, D i is the distance between the center of the ith equal partition and the building reference point, D max is the maximum reference distance of the building, and N is the number of equal partitions; If the brightness difference index is within a preset range, the energy consumption optimization information indicates that the illumination area is uniform, and an energy-saving control instruction is generated; If the brightness difference index is not within the preset range, the energy consumption optimization information indicates that the illumination area is uneven, and a dynamic adjustment instruction is generated.

8. The building energy consumption control method using illumination pattern changes according to claim 7, characterized in that: The step of generating an energy-saving control instruction or a dynamic adjustment instruction includes: When the energy consumption optimization information indicates that the illumination area is uniform, an energy-saving control instruction is generated to reduce the power of the lighting equipment or adjust the opening and closing angle of the sunshade device; When the energy consumption optimization information indicates that the illumination area is uneven, a dynamic adjustment instruction is generated for increasing the brightness of the lighting equipment or adjusting the angle of the sunshade device for a specific partition.

9. A building energy consumption control system using changes in illumination patterns, characterized in that: The building energy consumption control system includes: An image processing module is used to obtain indoor and outdoor lighting images of a building and generate a brightness distribution image corresponding to the lighting image; A feature positioning module, configured to obtain a feature point set of the illumination change area in the brightness distribution image based on a preset illumination feature positioning method; a feature correction module, configured to obtain distribution information of the feature point set to the stable illumination area in the brightness distribution image, and correct the feature point set according to the distribution information to obtain a corrected feature point set; A partition comparison module is used to perform partition comparison on the illumination area in the illumination image based on the corrected feature point set to obtain energy consumption optimization information of the illumination area.

10. The building energy consumption control system using light pattern changes according to claim 9, characterized in that: The partition comparison module further includes: a partitioning unit, configured to divide the illumination area in the illumination image into a preset number of equal partitions with the modified feature point set as the center; an analyzing unit, configured to obtain brightness distribution information of the equal subareas, and determine energy consumption optimization information of the illumination area according to differences in the brightness distribution information between the equal subareas; An instruction generation unit is used to generate energy-saving control instructions or dynamic adjustment instructions according to the energy consumption optimization information, and send them to the lighting equipment and shading devices in the building.

Images