AI dimming table lamp and intelligent dimming method thereof

By combining visual sensing and communication units with a pre-trained model, the brightness adjustment method of the AI-powered dimming desk lamp was developed, solving the problem of brightness adjustment when ambient light changes. This resulted in precise control of the lamp's brightness and optimal lighting effect.

CN120825847AActive Publication Date: 2025-10-21SHENZHEN TOJANE LIGHTING CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202511255513.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-04
Publication Date
2025-10-21
Estimated Expiration
2045-09-04

AI Technical Summary

Technical Problem

Existing desk lamps struggle to accurately adjust their brightness to meet users' lighting needs when ambient light changes, taking into account factors such as the light source, direction, size, and color of the desktop in different application environments.

Method used

An AI-powered dimming desk lamp is used, which combines a lighting unit, a visual sensing unit, and a communication unit. By acquiring a brightness image of the target lighting area, extracting brightness feature parameters, and using a pre-trained desk lamp control parameter generation model, the brightness of the desk lamp is adjusted to adapt to environmental changes.

Benefits of technology

It achieves precise control of the desk lamp's brightness, providing optimal lighting effects, adapting to different ambient light changes, and reducing eye discomfort.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120825847A_ABST
    Figure CN120825847A_ABST
Patent Text Reader

Abstract

The invention provides an AI dimming table lamp and an intelligent dimming method thereof, and the method comprises the steps: obtaining a brightness image of a target illumination region in a current dimming period in each dimming period, and extracting a brightness feature parameter of the target illumination region from the brightness image; judging whether the variation amplitude of the brightness characteristic parameter relative to the previous dimming period is greater than a preset threshold value or not, and when the variation amplitude of the brightness characteristic parameter relative to the previous dimming period is greater than the preset threshold value, generating an input data sequence containing the brightness characteristic parameter, the input data sequence is input into a pre-trained table lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, the output data sequence comprises brightness control parameters of the table lamp, the brightness of the table lamp is adjusted according to the brightness control parameters in the output data sequence, and the brightness of the table lamp is adjusted according to the brightness control parameters in the output data sequence. The illumination light of the table lamp can be accurately controlled so as to provide the optimal illumination effect.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of lighting control technology, and in particular to an AI dimming desk lamp and an intelligent dimming method thereof. Background Art

[0002] A desk lamp is used on a desk to provide optimized indoor lighting quality for reading and writing. While a main indoor light can also illuminate a desk, using a closer desk lamp provides higher illumination and more evenly distributed brightness directly on the desk surface, using less power. This helps prevent vision damage caused by insufficient brightness or glare from the main indoor light during prolonged use.

[0003] Indoor environments such as study rooms where desk lamps are used typically receive ambient light through windows. Fluctuations in ambient light intensity caused by the change of day and night, as well as climate, significantly impact the lighting performance of a desk lamp. For example, in conditions of strong ambient light, even direct sunlight, which is already too bright for reading or writing, and requires shading, leaving the desk lamp on can make the light on the desktop even more glaring. Another example is when ambient light is very low, such as at night or on rainy days, the entire room is dimmed, requiring the desk lamp's brightness to be adjusted to a brighter setting to provide sufficient light for reading and writing. Ambient light levels fluctuate frequently with time and weather. Furthermore, when the main indoor light is used for auxiliary lighting at night or during low daylight conditions, the brightness of different lamps can vary, making it difficult for the desk lamp to achieve the desired lighting effect.

[0004] To address the issue of fixed-brightness desk lamps not meeting user needs, some desk lamps on the market are equipped with integrated brightness sensors for detecting ambient light intensity. These sensors automatically adjust the brightness of the desk lamp based on the detected ambient light intensity, alleviating, to a certain extent, the eye discomfort caused by the mismatch between the desk lamp brightness and the ambient light intensity due to changes in ambient light. However, in different application environments, the source, direction, size of the light source, wall surface, and desk color of the ambient light can also have a significant impact on the lighting effect of the desk lamp. Simply adjusting the brightness of the desk lamp based on the light intensity detected by the brightness sensor often fails to meet the user's actual needs for desk lighting. Summary of the Invention

[0005] Based on the above problems, the present invention proposes an AI dimming desk lamp and an intelligent dimming method thereof, which can accurately control the lighting light of the desk lamp to provide the optimal lighting effect.

[0006] In view of this, a first aspect of the present invention provides an AI dimming desk lamp, comprising a lighting unit, a visual sensing unit, a communication unit, and a control unit, wherein the lighting unit, the visual sensing unit, and the communication unit are connected to the control unit, the visual sensing unit is configured to obtain a brightness image of a target lighting area, the communication unit is configured to establish a communication connection with a cloud server that provides a cloud-based brightness analysis model, and the control unit is configured to:

[0007] Configure the dimming cycle of the desk lamp;

[0008] Acquire a brightness image of the target lighting area in each dimming cycle, wherein the brightness image is an image composed of the brightness values ​​of the reflected light after the desk lamp light and the ambient light illuminate the target lighting area;

[0009] extracting brightness characteristic parameters of the target lighting area from the brightness image;

[0010] Determining whether a change in the brightness characteristic parameter relative to a previous dimming cycle is greater than a preset threshold;

[0011] When the change amplitude of the brightness characteristic parameter relative to the previous dimming cycle is greater than a preset threshold, generating an input data sequence including the brightness characteristic parameter;

[0012] Inputting the input data sequence into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, wherein the output data sequence includes the brightness control parameters of the desk lamp;

[0013] The brightness of the desk lamp is adjusted according to the brightness control parameters in the output data sequence.

[0014] A second aspect of the present invention provides an intelligent dimming method for an AI dimming desk lamp, comprising:

[0015] Configure the dimming cycle of the desk lamp;

[0016] Acquire a brightness image of the target lighting area in each dimming cycle, wherein the brightness image is an image composed of the brightness values ​​of the reflected light after the desk lamp light and the ambient light illuminate the target lighting area;

[0017] extracting brightness characteristic parameters of the target lighting area from the brightness image;

[0018] Determining whether a change in the brightness characteristic parameter relative to a previous dimming cycle is greater than a preset threshold;

[0019] When the change amplitude of the brightness characteristic parameter relative to the previous dimming cycle is greater than a preset threshold, generating an input data sequence including the brightness characteristic parameter;

[0020] Inputting the input data sequence into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, wherein the output data sequence includes the brightness control parameters of the desk lamp;

[0021] The brightness of the desk lamp is adjusted according to the brightness control parameters in the output data sequence.

[0022] Furthermore, the step of acquiring the brightness image of the target lighting area in the current dimming cycle in each dimming cycle specifically includes:

[0023] capturing a real-time image of the target lighting area through a visual sensing unit;

[0024] Performing edge recognition on the real-time image to obtain a plurality of color block areas, wherein the color block areas are areas surrounded by the same edge line and have one or more main colors;

[0025] Calculate the pixel area s of each color block area i , and the pixel distance d between the center coordinates of each color block area and the center coordinates of the target lighting area i , where i is 1 to n block A positive integer between block is the number of color block areas in the real-time image;

[0026] Calculate the lighting impact of each color block area:

[0027] I i =σ(d i )·s i ,

[0028] Where σ(d i ) is the distance d from the pixel i The distance coefficient associated with the size of

[0029] Generate a brightness image of the color block area whose lighting influence is greater than a preset influence threshold.

[0030] Furthermore, the step of generating a brightness image of a color block area whose lighting influence is greater than a preset influence threshold specifically includes:

[0031] Converting the real-time image into a target color mode having a brightness or luminance dimension;

[0032] Extracting the brightness value of each pixel point in the color block area from the real-time image of the target color mode to generate a brightness matrix having the same size as the real-time image;

[0033] Normalization is performed on the brightness values ​​in the brightness matrix to generate the brightness image.

[0034] Furthermore, the step of extracting the brightness characteristic parameters of the target lighting area from the brightness image specifically includes:

[0035] Counting the number of pixels corresponding to each brightness value in the brightness image;

[0036] generating a brightness-pixel number curve of the brightness image;

[0037] The curve characteristic parameters of the brightness-pixel number curve of the brightness image are extracted to calculate the brightness characteristic parameters of the target lighting area.

[0038] Furthermore, the step of extracting curve characteristic parameters of the brightness-pixel number curve of the brightness image to calculate the brightness characteristic parameters of the target lighting area specifically includes:

[0039] Get the peak coordinates of the brightness-pixel number curve [L peek , C peek ], where the peak brightness L peek is the brightness value with the largest number of pixels in the brightness image, and the peak pixel number C peek The brightness value in the brightness image is L peek The number of pixels;

[0040] Calculate the ratio of the number of peak pixels in the number of peak pixels in the brightness image:

[0041]

[0042] Wherein Hor is the pixel width of the brightness image, and Ver is the pixel height of the brightness image;

[0043] Calculate the median width W of the brightness-pixel number curve mid .

[0044] Furthermore, the median width W of the brightness-pixel number curve is calculated. mid The steps specifically include:

[0045] Draw a median line of the brightness-pixel number curve;

[0046] Count the number of intersections n between the median line and the brightness-pixel number curve inter ;

[0047] When n inter =1, obtain the brightness value L at the intersection inter ;

[0048] Calculate the slope k of the brightness-pixel number curve at the intersection inter ;

[0049] When k inter >0, the median width of the brightness-pixel number curve is configured as:

[0050] W mid =1-L inter ;

[0051] When k inter <0, the median width of the brightness-pixel number curve is configured as:

[0052] W mid =L inter .

[0053] Furthermore, when calculating the median width W of the brightness-pixel number curve, mid After the steps, it also includes:

[0054] Calculate the first brightness characteristic parameter of the target lighting area:

[0055]

[0056] Among them L peek,i is the peak brightness extracted from the brightness image of the i-th color block area;

[0057] Calculating a second brightness characteristic parameter of the target lighting area;

[0058]

[0059] where R peek,i is the percentage of peak pixels extracted from the brightness image of the i-th color block area;

[0060] Calculate the third brightness characteristic parameter of the target lighting area:

[0061]

[0062] Where W mid,i is the median width extracted from the brightness image of the i-th color block area;

[0063] The first brightness characteristic parameter P1, the second brightness characteristic parameter P2, and the third brightness characteristic parameter P3 are configured as the brightness characteristic parameters of the target lighting area.

[0064] Furthermore, the input data sequence further includes color characteristic parameters of the target lighting area, and after the step of converting the real-time image into a target color mode having a brightness or bright dimension, the step further includes:

[0065] Extracting the main color of the color block area from the real-time image of the target color mode, wherein the main color is the color with the largest number of pixels within a set color tolerance range in the color block area;

[0066] The main colors of one or more color block areas with the greatest lighting influence are configured as color characteristic parameters of the target lighting area.

[0067] Furthermore, the step of extracting the main color of the color block area from the real-time image of the target color mode specifically includes:

[0068] Get the pre-configured color tolerance distance in the target color mode;

[0069] Accumulating each color dimension of the target color mode according to a preset step size to traverse each color in the target color mode;

[0070] The traversed color is determined as the target color to perform the following steps:

[0071] Determine a target color set in a color space of the target color mode whose distance from the target color is less than or equal to the color tolerance distance;

[0072] Counting the number of pixels in the color block area that fall into the color set;

[0073] Establishing an association relationship between the number of pixels and the target color;

[0074] After the traversal is completed, the color with the largest number of associated pixels is determined as the main color of the color block area.

[0075] The present invention proposes an AI dimming desk lamp and an intelligent dimming method thereof. The method obtains a brightness image of a target lighting area in the current dimming cycle in each dimming cycle, extracts brightness characteristic parameters of the target lighting area from the brightness image, and determines whether the change amplitude of the brightness characteristic parameters relative to the previous dimming cycle is greater than a preset threshold. When the change amplitude of the brightness characteristic parameters relative to the previous dimming cycle is greater than the preset threshold, an input data sequence including the brightness characteristic parameters is generated, and the input data sequence is input into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence. The output data sequence includes the brightness control parameters of the desk lamp. The brightness of the desk lamp is adjusted according to the brightness control parameters in the output data sequence, which can accurately control the lighting light of the desk lamp to provide the optimal lighting effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 1 is a schematic diagram of an AI dimming desk lamp provided by one embodiment of the present invention;

[0077] Figure 2 This is a flowchart of an intelligent dimming method for an AI dimming desk lamp provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0078] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0079] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0080] In the description of the present invention, the term "plurality" refers to two or more. Unless otherwise specified, the terms "upper" and "lower" are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific manner. Therefore, they should not be construed as limiting the present invention. The terms "connected," "mounted," and "fixed," etc., should be interpreted broadly. For example, "connected" can refer to fixed, removable, or integral connections; directly or indirectly through an intermediary. A person of ordinary skill in the art will understand the specific meanings of these terms in the present invention based on the specific circumstances. Furthermore, the terms "first," "second," etc., etc., are used for descriptive purposes only and should not be construed to indicate or imply relative importance or to implicitly specify the number of the technical features indicated. Therefore, a feature designated "first," "second," etc., may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0081] Throughout this specification, terms such as "one embodiment," "some implementations," and "specific examples" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0082] An AI dimming desk lamp and an intelligent dimming method thereof provided according to some embodiments of the present invention will be described below with reference to the accompanying drawings.

[0083] like Figure 1 As shown, the first aspect of the present invention proposes an AI dimming desk lamp, comprising a lighting unit, a visual sensing unit, a communication unit and a control unit, wherein the lighting unit, the visual sensing unit and the communication unit are connected to the control unit, the visual sensing unit is configured to obtain a brightness image of a target lighting area, and the communication unit is configured to establish a communication connection with a cloud server that provides a cloud-based brightness analysis model.

[0084] In the technical solutions of some embodiments of the present invention, the lighting unit includes an LED and an LED driving unit connected to the LED, and the LED driving unit drives the LED to light up, turn off or adjust the brightness according to the control instructions of the control unit.

[0085] The visual sensing unit is a visual sensor such as a camera, which, under the control of the control unit, captures a real-time image of the target lighting area in each dimming cycle to extract a brightness image of the target lighting area from the real-time image.

[0086] The communication unit can be a wired communication unit, such as a communication unit connected to the Internet via an RS232 interface to establish a network communication connection with the outside, or a wireless communication unit using WIFI, Bluetooth, or other communication protocols. The desk lamp is connected to the Internet via the communication unit to establish a communication connection with the cloud server.

[0087] like Figure 2 As shown, the control unit is configured as follows:

[0088] Configure the dimming cycle of the desk lamp;

[0089] Acquire a brightness image of the target lighting area in each dimming cycle, wherein the brightness image is an image composed of the brightness values ​​of the reflected light after the desk lamp light and the ambient light illuminate the target lighting area;

[0090] extracting brightness characteristic parameters of the target lighting area from the brightness image;

[0091] Determining whether a change in the brightness characteristic parameter relative to a previous dimming cycle is greater than a preset threshold;

[0092] When the change amplitude of the brightness characteristic parameter relative to the previous dimming cycle is greater than a preset threshold, generating an input data sequence including the brightness characteristic parameter;

[0093] Inputting the input data sequence into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, wherein the output data sequence includes the brightness control parameters of the desk lamp;

[0094] The brightness of the desk lamp is adjusted according to the brightness control parameters in the output data sequence.

[0095] The dimming cycle is configured as a shorter time period to adapt to the dynamic changes in indoor ambient light. For example, the dimming cycle can be configured to 5 seconds, 10 seconds or other durations. The dimming cycle can be adaptively adjusted according to the changes in indoor light and the load status and response speed of the cloud server.

[0096] The target lighting area refers to the area illuminated by the lamp when it is placed on a desk or other location. For a non-adjustable desk lamp, the target lighting area is a fixed area, such as the desk surface, provided the lamp's position remains unchanged. For an adjustable desk lamp, such as one with a bendable support pole, the illumination direction can be adjusted freely, and the target lighting area will also change accordingly.

[0097] The brightness image is an image reflecting the real-time brightness of the target lighting area. The brightness image only contains the brightness information of the target lighting area, but not its color information. The brightness characteristic parameter is a characteristic parameter reflecting the brightness distribution of the target lighting area under the illumination of the desk lamp and ambient light.

[0098] The desk lamp control parameter generation model is an artificial intelligence model obtained by deep training using a large amount of sample data. Specifically, it can be any one of a convolutional neural network model, a recurrent neural network model, a generative adversarial network model, and a long short-term memory network model. Preferably, the desk lamp control parameter generation model is a long short-term memory network model. By utilizing the long-term memory capacity of the long short-term memory network model, it can memorize the changing patterns of indoor ambient light over a longer period, thereby more accurately generating the required desk lamp control parameters.

[0099] In the technical solution of the present invention, a large number of brightness characteristic parameters of the target lighting area of ​​the desk lamp are collected and organized as sample data in indoor environments at various times (including daytime and nighttime) and weather conditions, and the corresponding brightness control parameters are configured to perform supervised training on the desk lamp control parameter generation model. When the desk lamp control parameter generation model receives the brightness characteristic parameters in any indoor scene, it can generate desk lamp control parameters suitable for adjusting the light of the desk lamp in the indoor scene, so that the lighting conditions in the target lighting area are more in line with the needs of reading and writing.

[0100] Furthermore, in the step of acquiring the brightness image of the target lighting area in the current dimming cycle in each dimming cycle, the control unit is configured to:

[0101] capturing a real-time image of the target lighting area through a visual sensing unit;

[0102] Performing edge recognition on the real-time image to obtain a plurality of color block areas, wherein the color block areas are areas surrounded by the same edge line and have one or more main colors;

[0103] Calculate the pixel area s of each color block area i , and the pixel distance d between the center coordinates of each color block area and the center coordinates of the target lighting area i, where i is 1 to n block A positive integer between block is the number of color block areas in the real-time image;

[0104] Calculate the lighting impact of each color block area:

[0105] I i =σ(d i )·s i ,

[0106] Where σ(d i ) is the distance d from the pixel i The distance coefficient associated with the size of

[0107] Generate a brightness image of the color block area whose lighting influence is greater than a preset influence threshold.

[0108] In a specific application scenario, a desk surface is often filled with various colored items, such as books, mouse pads, and water cups. Each item typically has one or more primary colors. For example, an open book typically has a white background as its primary color, while more colorful items, such as color photos or posters, may have multiple primary colors. The above embodiment uses edge recognition to separate different items with edge lines, forming separate color blocks.

[0109] In some embodiments of the present invention, when the proportion of pixels of a color in a color block region to which it belongs exceeds a preset threshold, the color is referred to as the dominant color of the color block region. In other embodiments of the present invention, the colors that rank first in number of pixels within a color block region may be referred to as the dominant color of the color block region.

[0110] Furthermore, before the step of performing edge recognition on the real-time image to obtain a plurality of color block areas, the method further includes configuring a minimum pixel area of ​​the color block area;

[0111] The step of performing edge recognition on the real-time image to obtain a plurality of color block regions specifically includes configuring a region with a pixel area larger than the minimum pixel area as a color block region. It should be understood that the color block regions on the real-time image may have a containing and being contained relationship.

[0112] The distance coefficient σ(d i ) is the weight coefficient for calculating the lighting influence under different predefined pixel distances, and the distance coefficient σ(d i ) is the distance d from the pixel i The relationship function is inversely correlated, that is, the pixel distance d iThe larger the value is, the corresponding distance coefficient σ(d i ) is smaller. On the contrary, the pixel distance d i The smaller the distance coefficient σ(d i ) is larger. For example, the distance coefficient σ(d i ) can be the pixel distance d i The product of the reciprocal of and a specific constant, that is:

[0113]

[0114] Where σ0 is an empirical constant.

[0115] In the technical solution of the above embodiment, a brightness image is generated for each color block area of ​​the target lighting area whose lighting influence is greater than the preset influence threshold, that is, when there are n color block areas with lighting influence greater than the preset influence threshold in the target lighting area, m_block When the target lighting area has n m_block brightness images, and performing the brightness feature parameter extraction step for each brightness image.

[0116] Furthermore, in the step of generating a brightness image of a color block area whose lighting influence is greater than a preset influence threshold, the control unit is configured to:

[0117] Converting the real-time image into a target color mode having a brightness or luminance dimension;

[0118] Extracting the brightness value of each pixel point in the color block area from the real-time image of the target color mode to generate a brightness matrix having the same size as the real-time image;

[0119] Normalization is performed on the brightness values ​​in the brightness matrix to generate the brightness image.

[0120] Specifically, the target color mode may be any one of a YUV mode, a LAB mode, an HSB mode, or an HSL mode, wherein the Y value in the YUV mode, the L value in the LAB mode, the B value in the HSB mode, and the L value in the HSL mode are the brightness or lightness dimensions thereof. When the target color mode is the HSL mode, in the step of extracting the brightness value of each pixel from the real-time image of the target color mode, the lightness value L of each pixel is determined as the brightness value.

[0121] Furthermore, in the step of extracting the brightness value of each pixel point of the color block area from the real-time image of the target color mode to generate a brightness matrix having the same size as the real-time image, the control unit is configured to fill other pixel points outside the color block area to which the brightness image belongs with a filling number to expand the brightness matrix to the same pixel size as the real-time image, wherein the filling number can be the maximum boundary value or minimum boundary value of the brightness or luminance dimension in the corresponding color mode, so that it is normalized to 0 or 1 in subsequent normalization processing.

[0122] Furthermore, in the step of extracting the brightness characteristic parameters of the target lighting area from the brightness image, the control unit is configured to:

[0123] Counting the number of pixels corresponding to each brightness value in the brightness image;

[0124] generating a brightness-pixel number curve of the brightness image;

[0125] The curve characteristic parameters of the brightness-pixel number curve of the brightness image are extracted to calculate the brightness characteristic parameters of the target lighting area.

[0126] In the step of counting the number of pixels corresponding to each brightness value in the brightness image, the brightness image is a normalized brightness image. Therefore, the brightness values ​​referred to in this step are values ​​mapped to the interval [0, 1] with a specific accuracy. It should be noted that the brightness values ​​corresponding to the fill number are not within the statistical range.

[0127] Generating the brightness-pixel number curve of the brightness image specifically refers to fitting the discrete correspondence between the brightness value and the number of pixels in the brightness image into a relationship curve between the two, with the brightness value in the range of [0, 1] as the x-axis and the numerical range of the number of pixels as the y-axis.

[0128] Furthermore, in the step of extracting curve characteristic parameters of the brightness-pixel number curve of the brightness image to calculate the brightness characteristic parameters of the target lighting area, the control unit is configured to:

[0129] Get the peak coordinates of the brightness-pixel number curve [L peek , C peek ], where the peak brightness L peek is the brightness value with the largest number of pixels in the brightness image, and the peak pixel number C peek The brightness value in the brightness image is L peek The number of pixels;

[0130] Calculate the ratio of the number of peak pixels in the number of peak pixels in the brightness image:

[0131]

[0132] Wherein Hor is the pixel width of the brightness image, and Ver is the pixel height of the brightness image;

[0133] Calculate the median width W of the brightness-pixel number curve mid .

[0134] It should be noted that the peak brightness L peek It refers to the brightness value corresponding to the largest number of pixels in the brightness image, rather than the maximum brightness on the brightness image.

[0135] In the calculation formula of the peak pixel number ratio, Hor×Ver represents the total number of pixels in the brightness image.

[0136] Furthermore, when calculating the median width W of the brightness-pixel number curve, mid In the step, the control unit is configured to:

[0137] Draw a median line of the brightness-pixel number curve;

[0138] Count the number of intersections n between the median line and the brightness-pixel number curve inter ;

[0139] When n inter =1, obtain the brightness value L at the intersection inter ;

[0140] Calculate the slope k of the brightness-pixel number curve at the intersection inter ;

[0141] When k inter >0, the median width of the brightness-pixel number curve is configured as:

[0142] W mid =1-L inter ;

[0143] When k inter <0, the median width of the brightness-pixel number curve is configured as:

[0144] W mid =L inter .

[0145] The median line is in the brightness-pixel number two-dimensional orthogonal coordinate system, and the pixel number is The corresponding straight line, the median line is parallel to the brightness axis in the brightness-pixel number two-dimensional orthogonal coordinate system.

[0146] In the technical solution of the above embodiment, when n inter =1, it indicates that the median line and the brightness-pixel number curve have only one intersection point.

[0147] Furthermore, the number of intersections n between the median line and the brightness-pixel number curve is counted. inter After the steps, the control unit is configured to:

[0148] When n inter ≥2, obtain the first brightness value L1 of the minimum brightness value intersection point and the second brightness value L2 of the maximum brightness value intersection point;

[0149] Calculate the median width W of the brightness-pixel number curve mid =L2-L1.

[0150] Furthermore, in the extreme case where the median line has no intersection with the brightness-pixel number curve, the median width of the brightness-pixel number curve is configured to be 0.

[0151] Furthermore, when calculating the median width W of the brightness-pixel number curve, mid After the steps, the control unit is configured to:

[0152] Calculate the first brightness characteristic parameter of the target lighting area:

[0153]

[0154] Among them L peek,i is the peak brightness extracted from the brightness image of the i-th color block area;

[0155] Calculating a second brightness characteristic parameter of the target lighting area;

[0156]

[0157] where R peek,i is the percentage of peak pixels extracted from the brightness image of the i-th color block area;

[0158] Calculate the third brightness characteristic parameter of the target lighting area:

[0159]

[0160] Where W mid,i is the median width extracted from the brightness image of the i-th color block area;

[0161] The first brightness characteristic parameter P1, the second brightness characteristic parameter P2, and the third brightness characteristic parameter P3 are configured as the brightness characteristic parameters of the target lighting area.

[0162] Furthermore, in the step of determining whether the change amplitude of the brightness characteristic parameter relative to the previous dimming cycle is greater than a preset threshold, the control unit is configured to:

[0163] Obtaining a pre-configured threshold value of a change amplitude of each brightness characteristic parameter of the target lighting area;

[0164] Calculating the difference between the value of each brightness characteristic parameter of the target lighting area in the current dimming cycle and the value in the previous dimming cycle;

[0165] comparing the difference with the change magnitude threshold;

[0166] When the difference between any brightness characteristic parameter in two adjacent dimming cycles is greater than the corresponding change amplitude threshold, it is determined that the change amplitude of the brightness characteristic parameter of the target lighting area relative to the previous dimming cycle is greater than the preset threshold.

[0167] Furthermore, in the step of generating the input data sequence including the brightness characteristic parameters, the control unit is configured to:

[0168] Acquire working state parameters of the desk lamp, where the working state parameters include a switch state parameter S1 of the desk lamp, and also include a brightness parameter S2 and / or a power parameter S3 of the desk lamp;

[0169] The working state parameters of the desk lamp and the brightness characteristic parameters of the target lighting area are combined to form the input data sequence.

[0170] The switch state parameter S1 of the desk lamp can be configured to 0 and 1 to represent the on and off states of the desk lamp respectively. The brightness parameter S2 and power parameter S3 of the desk lamp are related to the working current of the desk lamp, and their numerical values ​​will change when the user adjusts the gear position of the desk lamp.

[0171] In the technical solutions of some embodiments of the present invention, the input data sequence can be expressed as:

[0172] {S1, S2, S3, P1, P2, P3}, wherein the brightness parameter S2 and the power parameter S3 of the desk lamp are both normalized values ​​after normalization processing.

[0173] Furthermore, the input data sequence further includes color characteristic parameters of the target lighting area. After the step of converting the real-time image into a target color mode having a brightness or bright dimension, the control unit is configured to:

[0174] Extracting the main color of the color block area from the real-time image of the target color mode, wherein the main color is the color with the largest number of pixels within a set color tolerance range in the color block area;

[0175] The main colors of one or more color block areas with the greatest lighting influence are configured as color characteristic parameters of the target lighting area.

[0176] The color space of each color mode has its own unique spatial structure. For example, the RGB color space is a cubic structure composed of red, green and blue color channels in an orthogonal relationship, while the HSB color space is a cylindrical structure with hue representing the circumferential dimension, saturation representing the radial dimension, and brightness representing the height dimension. The HSL color space is a biconical structure with hue representing the circumferential dimension, saturation representing the radial dimension, and brightness representing the height dimension. The color tolerance refers to the spatial distance in the color space of the corresponding color mode, that is, its three-dimensional coordinate distance. In a simpler calculation method, for any color mode, its color space can also be mapped to an orthogonal three-dimensional coordinate system, and the Euclidean distance of the color value of each color in the three-dimensional coordinate system is used as its color distance to determine the colors contained in its color tolerance range. The Euclidean distance between two colors is the square root of the sum of the squares of the differences between the color components of the two colors.

[0177] Taking the color characteristic parameters of the target lighting area as the main colors of the three color block areas with the greatest lighting influence as an example, the input data sequence can be expressed as:

[0178] {S1, S2, S3, C1, C2, C3, P1, P2, P3}, where C1, C2, and C3 are the moduli of the normalized vectors of the main colors of the three color block areas respectively.

[0179] In the technical solutions of some embodiments of the present invention, the output data sequence of the desk lamp control parameter generation model also includes the color temperature control parameters of the desk lamp. After the step of inputting the input data sequence into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, it also includes adjusting the color temperature of the desk lamp according to the color temperature control parameters in the output data sequence.

[0180] Furthermore, in the step of extracting the main color of the color block area from the real-time image of the target color mode, the control unit is configured to:

[0181] Get the pre-configured color tolerance distance in the target color mode;

[0182] Accumulating each color dimension of the target color mode according to a preset step size to traverse each color in the target color mode;

[0183] The traversed color is determined as the target color to perform the following steps:

[0184] Determine a target color set in a color space of the target color mode whose distance from the target color is less than or equal to the color tolerance distance;

[0185] Counting the number of pixels in the color block area that fall into the color set;

[0186] Establishing an association relationship between the number of pixels and the target color;

[0187] After the traversal is completed, the color with the largest number of associated pixels is determined as the main color of the color block area.

[0188] The color tolerance distance is the distance between colors in the color space corresponding to the target color mode, as reflected by the tolerance of the main color when counting the number of pixels of the main color in the color block area. The size of the color tolerance distance can be adaptively configured based on the color distribution of each color block area in the real-time image and the selected target color mode.

[0189] The color dimensions refer to parameters of each dimension constituting the target color model, such as the Y, U, and V dimensions in the YUV model, and the H, S, and L dimensions in the HSL model. In the step of accumulating each color dimension of the target color model according to a preset step size to traverse each color in the target color model, each dimension is typically accumulated sequentially using a minimum integer unit, such as 1 or 1 degree, as a step size until the accumulated value reaches an upper limit of the dimension.

[0190] like Figure 2 As shown, the second aspect of the present invention provides an intelligent dimming method for an AI dimming desk lamp, comprising:

[0191] Configure the dimming cycle of the desk lamp;

[0192] Acquire a brightness image of the target lighting area in each dimming cycle, wherein the brightness image is an image composed of the brightness values ​​of the reflected light after the desk lamp light and the ambient light illuminate the target lighting area;

[0193] extracting brightness characteristic parameters of the target lighting area from the brightness image;

[0194] Determining whether a change in the brightness characteristic parameter relative to a previous dimming cycle is greater than a preset threshold;

[0195] When the change amplitude of the brightness characteristic parameter relative to the previous dimming cycle is greater than a preset threshold, generating an input data sequence including the brightness characteristic parameter;

[0196] Inputting the input data sequence into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, wherein the output data sequence includes the brightness control parameters of the desk lamp;

[0197] The brightness of the desk lamp is adjusted according to the brightness control parameters in the output data sequence.

[0198] The dimming cycle is configured as a shorter time period to adapt to the dynamic changes in indoor ambient light. For example, the dimming cycle can be configured to 5 seconds, 10 seconds or other durations. The dimming cycle can be adaptively adjusted according to the changes in indoor light and the load status and response speed of the cloud server.

[0199] The target lighting area refers to the area illuminated by the lamp when it is placed on a desk or other location. For a non-adjustable desk lamp, the target lighting area is a fixed area, such as the desk surface, provided the lamp's position remains unchanged. For an adjustable desk lamp, such as one with a bendable support pole, the illumination direction can be adjusted freely, and the target lighting area will also change accordingly.

[0200] The brightness image is an image reflecting the real-time brightness of the target lighting area. The brightness image only contains the brightness information of the target lighting area, but not its color information. The brightness characteristic parameter is a characteristic parameter reflecting the brightness distribution of the target lighting area under the illumination of the desk lamp and ambient light.

[0201] The desk lamp control parameter generation model is an artificial intelligence model obtained by deep training using a large amount of sample data. Specifically, it can be any one of a convolutional neural network model, a recurrent neural network model, a generative adversarial network model, and a long short-term memory network model. Preferably, the desk lamp control parameter generation model is a long short-term memory network model. By utilizing the long-term memory capacity of the long short-term memory network model, it can memorize the changing patterns of indoor ambient light over a longer period, thereby more accurately generating the required desk lamp control parameters.

[0202] In the technical solution of the present invention, a large number of brightness characteristic parameters of the target lighting area of ​​the desk lamp are collected and organized as sample data in indoor environments at various times (including daytime and nighttime) and weather conditions, and the corresponding brightness control parameters are configured to perform supervised training on the desk lamp control parameter generation model. When the desk lamp control parameter generation model receives the brightness characteristic parameters in any indoor scene, it can generate desk lamp control parameters suitable for adjusting the light of the desk lamp in the indoor scene, so that the lighting conditions in the target lighting area are more in line with the needs of reading and writing.

[0203] Furthermore, the step of acquiring the brightness image of the target lighting area in the current dimming cycle in each dimming cycle specifically includes:

[0204] capturing a real-time image of the target lighting area through a visual sensing unit;

[0205] Performing edge recognition on the real-time image to obtain a plurality of color block areas, wherein the color block areas are areas surrounded by the same edge line and have one or more main colors;

[0206] Calculate the pixel area s of each color block area i , and the pixel distance d between the center coordinates of each color block area and the center coordinates of the target lighting area i , where i is 1 to n block A positive integer between block is the number of color block areas in the real-time image;

[0207] Calculate the lighting impact of each color block area:

[0208] I i =σ(d i )·s i ,

[0209] Where σ(d i ) is the distance d from the pixel i The distance coefficient associated with the size of

[0210] Generate a brightness image of the color block area whose lighting influence is greater than a preset influence threshold.

[0211] In a specific application scenario, a desk surface is often filled with various colored items, such as books, mouse pads, and water cups. Each item typically has one or more primary colors. For example, an open book typically has a white background as its primary color, while more colorful items, such as color photos or posters, may have multiple primary colors. The above embodiment uses edge recognition to separate different items with edge lines, forming separate color blocks.

[0212] In some embodiments of the present invention, when the proportion of pixels of a color in a color block region to which it belongs exceeds a preset threshold, the color is referred to as the dominant color of the color block region. In other embodiments of the present invention, the colors that rank first in number of pixels within a color block region may be referred to as the dominant color of the color block region.

[0213] Furthermore, before the step of performing edge recognition on the real-time image to obtain a plurality of color block areas, the method further includes configuring a minimum pixel area of ​​the color block area;

[0214] The step of performing edge recognition on the real-time image to obtain a plurality of color block regions specifically includes configuring a region with a pixel area larger than the minimum pixel area as a color block region. It should be understood that the color block regions on the real-time image may have a containing and being contained relationship.

[0215] The distance coefficient σ(d i ) is the weight coefficient for calculating the lighting influence under different predefined pixel distances, and the distance coefficient σ(d i ) is the distance d from the pixel i The relationship function is inversely correlated, that is, the pixel distance d i The larger the value is, the corresponding distance coefficient σ(d i ) is smaller. On the contrary, the pixel distance d i The smaller the distance coefficient σ(d i ) is larger. For example, the distance coefficient σ(d i ) can be the pixel distance d i The product of the reciprocal of and a specific constant, that is:

[0216]

[0217] Where σ0 is an empirical constant.

[0218] In the technical solution of the above embodiment, a brightness image is generated for each color block area of ​​the target lighting area whose lighting influence is greater than the preset influence threshold, that is, when there are n color block areas with lighting influence greater than the preset influence threshold in the target lighting area, m_block When the target lighting area has n m_block brightness images, and performing the brightness feature parameter extraction step for each brightness image.

[0219] Furthermore, the step of generating a brightness image of a color block area whose lighting influence is greater than a preset influence threshold specifically includes:

[0220] Converting the real-time image into a target color mode having a brightness or luminance dimension;

[0221] Extracting the brightness value of each pixel point in the color block area from the real-time image of the target color mode to generate a brightness matrix having the same size as the real-time image;

[0222] Normalization is performed on the brightness values ​​in the brightness matrix to generate the brightness image.

[0223] Specifically, the target color mode may be any one of a YUV mode, a LAB mode, an HSB mode, or an HSL mode, wherein the Y value in the YUV mode, the L value in the LAB mode, the B value in the HSB mode, and the L value in the HSL mode are the brightness or lightness dimensions thereof. When the target color mode is the HSL mode, in the step of extracting the brightness value of each pixel from the real-time image of the target color mode, the lightness value L of each pixel is determined as the brightness value.

[0224] Furthermore, in the step of extracting the brightness value of each pixel point of the color block area from the real-time image of the target color mode to generate a brightness matrix having the same size as the real-time image, other pixel points outside the color block area to which the brightness image belongs are filled with a filling number to expand the brightness matrix to the same pixel size as the real-time image, wherein the filling number can be the maximum boundary value or minimum boundary value of the brightness or luminance dimension in the corresponding color mode, so that it is normalized to 0 or 1 in the subsequent normalization processing.

[0225] Furthermore, the step of extracting the brightness characteristic parameters of the target lighting area from the brightness image specifically includes:

[0226] Counting the number of pixels corresponding to each brightness value in the brightness image;

[0227] generating a brightness-pixel number curve of the brightness image;

[0228] The curve characteristic parameters of the brightness-pixel number curve of the brightness image are extracted to calculate the brightness characteristic parameters of the target lighting area.

[0229] In the step of counting the number of pixels corresponding to each brightness value in the brightness image, the brightness image is a normalized brightness image. Therefore, the brightness values ​​referred to in this step are values ​​mapped to the interval [0, 1] with a specific accuracy. It should be noted that the brightness values ​​corresponding to the fill number are not within the statistical range.

[0230] Generating the brightness-pixel number curve of the brightness image specifically refers to fitting the discrete correspondence between the brightness value and the number of pixels in the brightness image into a relationship curve between the two, with the brightness value in the range of [0, 1] as the x-axis and the numerical range of the number of pixels as the y-axis.

[0231] Furthermore, the step of extracting curve characteristic parameters of the brightness-pixel number curve of the brightness image to calculate the brightness characteristic parameters of the target lighting area specifically includes:

[0232] Get the peak coordinates of the brightness-pixel number curve [L peek , C peek ], where the peak brightness L peek is the brightness value with the largest number of pixels in the brightness image, and the peak pixel number C peek The brightness value in the brightness image is L peek The number of pixels;

[0233] Calculate the ratio of the number of peak pixels in the number of peak pixels in the brightness image:

[0234]

[0235] Wherein Hor is the pixel width of the brightness image, and Ver is the pixel height of the brightness image;

[0236] Calculate the median width W of the brightness-pixel number curve mid .

[0237] It should be noted that the peak brightness L peek It refers to the brightness value corresponding to the largest number of pixels in the brightness image, rather than the maximum brightness on the brightness image.

[0238] In the calculation formula of the peak pixel number ratio, Hor×Ver represents the total number of pixels in the brightness image.

[0239] Furthermore, the median width W of the brightness-pixel number curve is calculated. mid The steps specifically include:

[0240] Draw a median line of the brightness-pixel number curve;

[0241] Count the number of intersections n between the median line and the brightness-pixel number curve inter ;

[0242] When n inter =1, obtain the brightness value L at the intersection inter ;

[0243] Calculate the slope k of the brightness-pixel number curve at the intersection inter ;

[0244] When k inter >0, the median width of the brightness-pixel number curve is configured as:

[0245] W mid =1-L inter ;

[0246] When k inter <0, the median width of the brightness-pixel number curve is configured as:

[0247] W mid =L inter .

[0248] The median line is in the brightness-pixel number two-dimensional orthogonal coordinate system, and the pixel number is The corresponding straight line, the median line is parallel to the brightness axis in the brightness-pixel number two-dimensional orthogonal coordinate system.

[0249] In the technical solution of the above embodiment, when n inter =0, it indicates that the median line and the brightness-pixel number curve have only one intersection point.

[0250] Furthermore, the number of intersections n between the median line and the brightness-pixel number curve is counted. inter After the steps, it also includes:

[0251] When n inter ≥2, obtain the first brightness value L1 of the minimum brightness value intersection point and the second brightness value L2 of the maximum brightness value intersection point;

[0252] Calculate the median width W of the brightness-pixel number curve mid =L2-L1.

[0253] Furthermore, in the extreme case where the median line has no intersection with the brightness-pixel number curve, the median width of the brightness-pixel number curve is configured to be 0.

[0254] Furthermore, when calculating the median width W of the brightness-pixel number curve, mid After the steps, it also includes:

[0255] Calculate the first brightness characteristic parameter of the target lighting area:

[0256]

[0257] Among them L peek,iis the peak brightness extracted from the brightness image of the i-th color block area;

[0258] Calculating a second brightness characteristic parameter of the target lighting area;

[0259]

[0260] where R peek,i is the percentage of peak pixels extracted from the brightness image of the i-th color block area;

[0261] Calculate the third brightness characteristic parameter of the target lighting area:

[0262]

[0263] Where W mid,i is the median width extracted from the brightness image of the i-th color block area;

[0264] The first brightness characteristic parameter P1, the second brightness characteristic parameter P2, and the third brightness characteristic parameter P3 are configured as the brightness characteristic parameters of the target lighting area.

[0265] Furthermore, the step of determining whether the change amplitude of the brightness characteristic parameter relative to the previous dimming cycle is greater than a preset threshold specifically includes:

[0266] Obtaining a pre-configured threshold value of a change amplitude of each brightness characteristic parameter of the target lighting area;

[0267] Calculating the difference between the value of each brightness characteristic parameter of the target lighting area in the current dimming cycle and the value in the previous dimming cycle;

[0268] comparing the difference with the change magnitude threshold;

[0269] When the difference between any brightness characteristic parameter in two adjacent dimming cycles is greater than the corresponding change amplitude threshold, it is determined that the change amplitude of the brightness characteristic parameter of the target lighting area relative to the previous dimming cycle is greater than the preset threshold.

[0270] Furthermore, the step of generating an input data sequence including the brightness characteristic parameters specifically includes:

[0271] Acquire working state parameters of the desk lamp, where the working state parameters include a switch state parameter S1 of the desk lamp, and also include a brightness parameter S2 and / or a power parameter S3 of the desk lamp;

[0272] The working state parameters of the desk lamp and the brightness characteristic parameters of the target lighting area are combined to form the input data sequence.

[0273] The switch state parameter S1 of the desk lamp can be configured to 0 and 1 to represent the on and off states of the desk lamp respectively. The brightness parameter S2 and power parameter S3 of the desk lamp are related to the working current of the desk lamp, and their numerical values ​​will change when the user adjusts the gear position of the desk lamp.

[0274] In the technical solutions of some embodiments of the present invention, the input data sequence can be expressed as:

[0275] {S1, S2, S3, P1, P2, P3}, wherein the brightness parameter S2 and the power parameter S3 of the desk lamp are both normalized values ​​after normalization processing.

[0276] Furthermore, the input data sequence further includes color characteristic parameters of the target lighting area, and after the step of converting the real-time image into a target color mode having a brightness or bright dimension, the step further includes:

[0277] Extracting the main color of the color block area from the real-time image of the target color mode, wherein the main color is the color with the largest number of pixels within a set color tolerance range in the color block area;

[0278] The main colors of one or more color block areas with the greatest lighting influence are configured as color characteristic parameters of the target lighting area.

[0279] The color space of each color mode has its own unique spatial structure. For example, the RGB color space is a cubic structure composed of red, green and blue color channels in an orthogonal relationship, while the HSB color space is a cylindrical structure with hue representing the circumferential dimension, saturation representing the radial dimension, and brightness representing the height dimension. The HSL color space is a biconical structure with hue representing the circumferential dimension, saturation representing the radial dimension, and brightness representing the height dimension. The color tolerance refers to the spatial distance in the color space of the corresponding color mode, that is, its three-dimensional coordinate distance. In a simpler calculation method, for any color mode, its color space can also be mapped to an orthogonal three-dimensional coordinate system, and the Euclidean distance of the color value of each color in the three-dimensional coordinate system is used as its color distance to determine the colors contained in its color tolerance range. The Euclidean distance between two colors is the square root of the sum of the squares of the differences between the color components of the two colors.

[0280] Taking the color characteristic parameters of the target lighting area as the main colors of the three color block areas with the greatest lighting influence as an example, the input data sequence can be expressed as:

[0281] {S1, S2, S3, C1, C2, C3, P1, P2, P3}, where C1, C2, and C3 are the moduli of the normalized vectors of the main colors of the three color block areas respectively.

[0282] In the technical solutions of some embodiments of the present invention, the output data sequence of the desk lamp control parameter generation model also includes the color temperature control parameters of the desk lamp. After the step of inputting the input data sequence into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, it also includes adjusting the color temperature of the desk lamp according to the color temperature control parameters in the output data sequence.

[0283] Furthermore, the step of extracting the main color of the color block area from the real-time image of the target color mode specifically includes:

[0284] Get the pre-configured color tolerance distance in the target color mode;

[0285] Accumulating each color dimension of the target color mode according to a preset step size to traverse each color in the target color mode;

[0286] The traversed color is determined as the target color to perform the following steps:

[0287] Determine a target color set in a color space of the target color mode whose distance from the target color is less than or equal to the color tolerance distance;

[0288] Counting the number of pixels in the color block area that fall into the color set;

[0289] Establishing an association relationship between the number of pixels and the target color;

[0290] After the traversal is completed, the color with the largest number of associated pixels is determined as the main color of the color block area.

[0291] The color tolerance distance is the distance between colors in the color space corresponding to the target color mode, as reflected by the tolerance of the main color when counting the number of pixels of the main color in the color block area. The size of the color tolerance distance can be adaptively configured based on the color distribution of each color block area in the real-time image and the selected target color mode.

[0292] The color dimensions refer to parameters of each dimension constituting the target color model, such as the Y, U, and V dimensions in the YUV model, and the H, S, and L dimensions in the HSL model. In the step of accumulating each color dimension of the target color model according to a preset step size to traverse each color in the target color model, each dimension is typically accumulated sequentially using a minimum integer unit, such as 1 or 1 degree, as a step size until the accumulated value reaches an upper limit of the dimension.

[0293] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0294] While embodiments of the present invention have been described above, these embodiments do not exhaustively describe all details and do not limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the above description. These embodiments are selected and described in detail in this specification in order to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better utilize the present invention and its modifications. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An AI dimming desk lamp, characterized in that: The system comprises an illumination unit, a visual sensing unit, a communication unit, and a control unit. The illumination unit, the visual sensing unit, and the communication unit are connected to the control unit. The visual sensing unit is configured to obtain a brightness image of a target illumination area. The communication unit is configured to establish a communication connection with a cloud server that provides a cloud-based brightness analysis model. The control unit is configured to: Configure the dimming cycle of the desk lamp; Acquire a brightness image of the target lighting area in each dimming cycle, wherein the brightness image is an image composed of the brightness values ​​of the reflected light after the desk lamp light and the ambient light illuminate the target lighting area; extracting brightness characteristic parameters of the target lighting area from the brightness image; Determining whether a change in the brightness characteristic parameter relative to a previous dimming cycle is greater than a preset threshold; When the change amplitude of the brightness characteristic parameter relative to the previous dimming cycle is greater than a preset threshold, generating an input data sequence including the brightness characteristic parameter; Inputting the input data sequence into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, wherein the output data sequence includes the brightness control parameters of the desk lamp; The brightness of the desk lamp is adjusted according to the brightness control parameters in the output data sequence.

2. An intelligent dimming method for an AI dimming desk lamp, characterized in that: include: Configure the dimming cycle of the desk lamp; Acquire a brightness image of the target lighting area in each dimming cycle, wherein the brightness image is an image composed of the brightness values ​​of the reflected light after the desk lamp light and the ambient light illuminate the target lighting area; extracting brightness characteristic parameters of the target lighting area from the brightness image; Determining whether a change in the brightness characteristic parameter relative to a previous dimming cycle is greater than a preset threshold; When the change amplitude of the brightness characteristic parameter relative to the previous dimming cycle is greater than a preset threshold, generating an input data sequence including the brightness characteristic parameter; Inputting the input data sequence into a pre-trained desk lamp control parameter generation model to obtain an output data sequence corresponding to the input data sequence, wherein the output data sequence includes the brightness control parameters of the desk lamp; The brightness of the desk lamp is adjusted according to the brightness control parameters in the output data sequence.

3. The intelligent dimming method of the AI ​​dimming desk lamp according to claim 2, characterized in that: The step of acquiring the brightness image of the target lighting area in the current dimming cycle in each dimming cycle specifically includes: capturing a real-time image of the target lighting area through a visual sensing unit; Performing edge recognition on the real-time image to obtain a plurality of color block areas, wherein the color block areas are areas surrounded by the same edge line and have one or more main colors; Calculate the pixel area s of each color block area i , and the pixel distance d between the center coordinates of each color block area and the center coordinates of the target lighting area i , where i is 1 to n block A positive integer between block is the number of color block areas in the real-time image; Calculate the lighting impact of each color block area: I i =σ(d i )·s i , Where σ(d i ) is the distance d from the pixel i The distance coefficient associated with the size of Generate a brightness image of the color block area whose lighting influence is greater than a preset influence threshold.

4. The intelligent dimming method of the AI ​​dimming desk lamp according to claim 3, characterized in that: The step of generating a brightness image of a color block area whose lighting influence is greater than a preset influence threshold specifically includes: Converting the real-time image into a target color mode having a brightness or luminance dimension; Extracting the brightness value of each pixel point in the color block area from the real-time image of the target color mode to generate a brightness matrix having the same size as the real-time image; Normalization is performed on the brightness values ​​in the brightness matrix to generate the brightness image.

5. The intelligent dimming method of the AI ​​dimming desk lamp according to claim 4, characterized in that: The step of extracting the brightness characteristic parameters of the target lighting area from the brightness image specifically includes: Counting the number of pixels corresponding to each brightness value in the brightness image; generating a brightness-pixel number curve of the brightness image; The curve characteristic parameters of the brightness-pixel number curve of the brightness image are extracted to calculate the brightness characteristic parameters of the target lighting area.

6. The intelligent dimming method of the AI ​​dimming desk lamp according to claim 5, characterized in that: The step of extracting curve characteristic parameters of the brightness-pixel number curve of the brightness image and calculating the brightness characteristic parameters of the target lighting area specifically includes: Get the peak coordinates of the brightness-pixel number curve [L peek , C peek ], where the peak brightness L peek is the brightness value with the largest number of pixels in the brightness image, and the peak pixel number C peek The brightness value in the brightness image is L peek The number of pixels; Calculate the ratio of the number of peak pixels in the number of peak pixels in the brightness image: Wherein Hor is the pixel width of the brightness image, and Ver is the pixel height of the brightness image; Calculate the median width W of the brightness-pixel number curve mid .

7. The intelligent dimming method of the AI ​​dimming desk lamp according to claim 6, characterized in that: Calculate the median width W of the brightness-pixel number curve mid The steps specifically include: Draw a median line of the brightness-pixel number curve; Count the number of intersections n between the median line and the brightness-pixel number curve inter ; When n inter =1, obtain the brightness value L at the intersection inter ; Calculate the slope k of the brightness-pixel number curve at the intersection inter ; When k inter >0, the median width of the brightness-pixel number curve is configured as: W mid =1-L inter ; When k inter When <0, the median width of the brightness-pixel number curve is configured as: W mid =L inter 。 8. The intelligent dimming method of the AI ​​dimming desk lamp according to claim 6, characterized in that: When calculating the median width W of the brightness-pixel number curve mid After the steps, it also includes: Calculate the first brightness characteristic parameter of the target lighting area: Among them L peek,i is the peak brightness extracted from the brightness image of the i-th color block area; Calculating a second brightness characteristic parameter of the target lighting area; where R peek,i is the percentage of peak pixels extracted from the brightness image of the i-th color block area; Calculate the third brightness characteristic parameter of the target lighting area: Where W mid,i is the median width extracted from the brightness image of the i-th color block area; The first brightness characteristic parameter P1, the second brightness characteristic parameter P2, and the third brightness characteristic parameter P3 are configured as the brightness characteristic parameters of the target lighting area.

9. The intelligent dimming method of the AI ​​dimming desk lamp according to claim 4, characterized in that: The input data sequence further includes color characteristic parameters of the target lighting area, and after the step of converting the real-time image into a target color mode having a brightness or bright dimension, the step further includes: Extracting the main color of the color block area from the real-time image of the target color mode, wherein the main color is the color with the largest number of pixels within a set color tolerance range in the color block area; The main colors of one or more color block areas with the greatest lighting influence are configured as color characteristic parameters of the target lighting area.

10. The intelligent dimming method of the AI ​​dimming desk lamp according to claim 9, characterized in that: The step of extracting the main color of the color block area from the real-time image of the target color mode specifically includes: Get the pre-configured color tolerance distance in the target color mode; Accumulating each color dimension of the target color mode according to a preset step size to traverse each color in the target color mode; The traversed color is determined as the target color to perform the following steps: Determine a target color set in a color space of the target color mode whose distance from the target color is less than or equal to the color tolerance distance; Counting the number of pixels in the color block area that fall into the color set; Establishing an association relationship between the number of pixels and the target color; After the traversal is completed, the color with the largest number of associated pixels is determined as the main color of the color block area.

Citation Information

Patent Citations

  • Electrooptic device, driving circuit, and electronic device

    CN101025492A

  • Adaptive brightness control method and system for projection equipment

    CN113727084A

  • Intelligent lamp control method and control system based on environment image

    CN114549864A

  • Urban light optimization model training system and method

    CN117094230A

  • Image adjuster and image adjusting method and program

    US20140078247A1