A low-light image enhancement method for construction sites

Abstract

The present invention relates to the field of computer vision, and in particular to a weak-light image enhancement method for a construction site. The present invention controls a mobile image acquisition unit to acquire a target area under conditions that do not meet weak-light conditions, so as to obtain a number of first image data covering the target area; extracts first image data for different sub-target areas, identifies object salient features, calculates object salient feature characterization parameters, and determines feature salient categories of the sub-target areas; determines image acquisition accuracy for acquiring the sub-target areas according to the feature salient categories of the sub-target areas, acquires the target areas under weak-light conditions with corresponding image acquisition accuracy, and acquires a number of second image data covering the target areas; processes the second image data of the sub-target areas according to the feature salient categories of the sub-target areas; the present invention processes massive image data in a targeted manner, and reduces computing power consumption caused by analyzing image data while ensuring reliability.

Description

Technical Field

[0001] The present invention relates to the field of computer vision, and in particular to a weak-light image enhancement method for a construction site. Background Art

[0002] Target detection at construction sites at night faces problems such as poor image quality and low target-background differentiation. Existing night image enhancement methods usually rely on high-cost hardware equipment, such as night vision goggles and infrared cameras. Although target detection methods in the field of computer vision have achieved remarkable results under daytime conditions, they still suffer from performance degradation under nighttime conditions.

[0003] Chinese Patent Publication No.: CN109886901A, discloses a night image enhancement method based on multi-channel decomposition, which is applied to the field of image processing technology. The invention first decomposes the night image to be processed into a base layer image and a detail layer image according to the estimation of the global noise level of the night image to be processed; then performs brightness adaptation calculation on the base layer image; secondly, corrects the color of the base layer brightness adaptation image; again, performs edge protection and noise suppression on the detail layer image; finally, fuses the color-corrected base layer image with the edge-protected and noise-suppressed detail layer image to obtain an enhanced night image.

[0004] However, there are still the following problems in the prior art:

[0005] In the prior art, when facing the weak-light environment monitoring of a large-scale construction site, due to the wide target area, the acquisition end will collect a huge amount of image data. In addition, since the images corresponding to the construction site may include several featureless areas without valuable information, such as repeated empty areas, etc., the traversal enhancement analysis method is used to analyze the massive image data, which consumes a lot of computing power. Summary of the invention

[0006] To this end, the present invention provides a low-light image enhancement method for construction sites, so as to overcome the problem that in the prior art, when facing low-light environment monitoring of large-scale construction sites, due to the wide target area, the acquisition end will collect a large amount of image data, and since the image corresponding to the construction site may include several featureless areas without valuable information, the traversal enhancement analysis method is used to analyze the massive image data, which consumes a lot of computing power.

[0007] To achieve the above object, the present invention provides a weak-light image enhancement method for a construction site, comprising:

[0008] Step S1, controlling the mobile image acquisition unit to acquire a target area under a weak light condition to obtain a plurality of first image data covering the target area;

[0009] Step S2, extracting first image data for different sub-target areas, identifying object salient features in the first image data, and calculating object salient feature characterization parameters based on the object salient features to determine feature salient categories of the sub-target areas, wherein the object salient features include the number of object contours and the texture density in the object contours;

[0010] Step S3, determining the image acquisition accuracy for the sub-target area according to the feature highlighting category of the sub-target area, acquiring the target area with the corresponding image acquisition accuracy under weak light conditions, and obtaining a plurality of second image data covering the target area;

[0011] Step S4, processing the second image data of the sub-target area according to the feature highlighting category of the sub-target area, including:

[0012] Determine a feature area according to the first image data of the sub-target area, perform image enhancement and verification on the feature area in the second image data of the sub-target area, and transmit the verified second image data to the supervision end;

[0013] Alternatively, global image enhancement is performed on the second image data of the sub-target area and the image is transmitted to the monitoring end.

[0014] Furthermore, in step S2, the object salient feature characterization parameter is calculated based on the object salient feature according to formula (1), including:

[0015]

[0016] In formula (1), R represents the parameter representing the object's prominent features, fe(x) represents the feature prominent function, N represents the number of object contours, N0 represents the number threshold, M represents the texture density in the object contour, M0 represents the texture density threshold, α represents the number weight coefficient of the object contour, β represents the texture density weight coefficient in the object contour, and e represents a natural constant.

[0017] Furthermore, in step S2, determining the feature highlight category of the sub-target area includes:

[0018] If the object salient feature representation parameter is greater than or equal to the object salient feature representation parameter threshold, the feature salient feature of the sub-target area is determined to be a strong feature salient category;

[0019] If the object salient feature representation parameter is less than the object salient feature representation parameter threshold, the feature salient feature of the sub-target area is determined to be a weak feature salient category.

[0020] Furthermore, in step S3, the image acquisition accuracy for the sub-target area is determined according to the feature highlighting category of the sub-target area, including:

[0021] If the feature highlighting category of the sub-target area is a weak feature highlighting category, it is determined that the reference image acquisition accuracy is used when acquiring the sub-target area;

[0022] If the feature highlighting category of the sub-target area is a strong feature highlighting category, the base image acquisition accuracy is increased when acquiring the sub-target area.

[0023] Furthermore, in step S4, the second image data of the sub-target area is processed according to the feature highlighting category of the sub-target area, including:

[0024] If the feature highlighting of the sub-target area is a weak feature highlighting category, determining the feature area according to the first image data of the sub-target area, performing image enhancement and verification on the feature area in the second image data of the sub-target area, and transmitting the verified second image data to the supervision end;

[0025] If the feature highlighting of the sub-target area is a strong feature highlighting category, global image enhancement is performed on the second image data of the sub-target area and transmitted to the supervision end.

[0026] Furthermore, in step S4, the process of determining the feature area according to the first image data of the sub-target area includes:

[0027] Dividing the first image data into a plurality of image regions and setting sequence numbers;

[0028] Determine an image region that meets a feature screening condition and an adjacent image region of the image region as a feature region;

[0029] The feature screening condition is that there is an object contour in the image area and the area of ​​the object contour must be greater than a predetermined area threshold.

[0030] Furthermore, in step S4, the process of performing image enhancement on the feature area in the second image data of the sub-target area includes:

[0031] Dividing the second image data into a plurality of image regions and setting sequence numbers;

[0032] The image region to be enhanced in the second image data is determined based on the sequence number of each image region in the determined feature region.

[0033] Furthermore, in step S4, the verification process includes:

[0034] determining a sub-target area corresponding to the second image data;

[0035] Extracting first image data corresponding to the sub-target area;

[0036] Determine the degree of overlap of the object contour of the first image data and the second image data;

[0037] If the object contour coincidence is greater than a predetermined coincidence threshold, it is determined that the second image data meets the verification standard.

[0038] Furthermore, in step S1, the mobile image acquisition unit moves along a predetermined route and acquires image data at a fixed acquisition point.

[0039] Furthermore, the low-light condition is that the current ambient light intensity is less than a predetermined light intensity threshold.

[0040] Compared with the prior art, the present invention controls the mobile image acquisition unit to acquire the target area under the condition that does not meet the weak light condition, so as to obtain a number of first image data covering the target area; extracts the first image data for different sub-target areas, identifies the prominent features of the object, and calculates the characteristic parameters of the object prominent features to determine the feature highlighting category of the sub-target area; determines the image acquisition accuracy for acquiring the sub-target area according to the feature highlighting category of the sub-target area, acquires the target area with the corresponding image acquisition accuracy under the weak light condition, and obtains a number of second image data covering the target area; processes the second image data of the sub-target area according to the feature highlighting category of the sub-target area; the present invention processes the massive image data in a targeted manner, and reduces the computing power consumption brought by analyzing the image data while ensuring reliability.

[0041] In particular, the present invention performs unified image acquisition of each target area of ​​the construction site in advance without meeting weak light conditions to obtain first image data as a reference. The first image data is collected without meeting weak light conditions, is relatively clear, and has obvious features. In actual situations, when performing image acquisition on a large-scale construction site, it is important to monitor feature areas, such as material accumulation areas, construction areas, etc. However, for large-scale construction areas, there may also be a large number of featureless areas, such as vacant land, which is reflected in the lack of obvious features and valuable information in the image data, and the image data is single, which is not easy to perform feature extraction and analysis, and the analysis value is low. Therefore, the present invention considers pre-determining the object salient features of each target area through the first image data, and correspondingly calculating the object salient feature characterization parameters. Since the object salient features include the number of object contours and the texture density in the object contours, the calculated object salient feature characterization parameters can characterize the value information contained in different target areas, and provide support for the subsequent adaptive analysis of the second image data. Furthermore, the massive image data is processed in a targeted manner, and the computing power consumption brought by the analysis of the image data is reduced while ensuring reliability.

[0042] In particular, the present invention divides the feature highlighting categories of the sub-target area based on the first image data, and then, different acquisition accuracies are used when acquiring the second image data subsequently. For the strong feature highlighting area, since the feature area has more features, it is collected with higher accuracy, which is convenient for the subsequent global image enhancement of the collected second highlighting data for the strong feature highlighting area. For the weak feature highlighting area, it is considered that there may be subdivided areas without obvious features and valuable information. Therefore, for the image data collected for the weak feature highlighting area, the feature area with features is determined, and only the feature area is enhanced to avoid feature omission, thereby reducing the computing power consumption brought by analyzing the image data while ensuring reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A schematic diagram of the steps of a weak-light image enhancement method for a construction site according to an embodiment of the invention;

[0044] Figure 2 A logical decision diagram for determining the feature highlighting category of a sub-target area in an embodiment of the invention;

[0045] Figure 3 A logic decision diagram for determining the image acquisition accuracy for the sub-target area for the embodiment of the invention;

[0046] Figure 4 This is a logic decision diagram for processing the second image data of the sub-object area according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0048] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0049] See also Figures 1 to 4 As shown, Figure 1 Schematic diagram of the steps of a weak-light image enhancement method for a construction site according to an embodiment of the present invention. Figure 2 A logical decision diagram for determining the image acquisition accuracy for the sub-target area for the embodiment of the invention, Figure 3 A logical decision diagram for determining the feature highlighting category of a sub-target area in an embodiment of the invention, Figure 4 The logic decision diagram for processing the second image data of the sub-target area according to the embodiment of the invention is as follows: the weak-light image enhancement method for a construction site according to the embodiment of the invention comprises:

[0050] Step S1, controlling the mobile image acquisition unit to acquire a target area under a weak light condition to obtain a plurality of first image data covering the target area;

[0051] Step S2, extracting first image data for different sub-target areas, identifying object salient features in the first image data, and calculating object salient feature characterization parameters based on the object salient features to determine feature salient categories of the sub-target areas, wherein the object salient features include the number of object contours and the texture density in the object contours;

[0052] Step S3, determining the image acquisition accuracy for the sub-target area according to the feature highlighting category of the sub-target area, acquiring the target area with the corresponding image acquisition accuracy under weak light conditions, and obtaining a plurality of second image data covering the target area;

[0053] Step S4, processing the second image data of the sub-target area according to the feature highlighting category of the sub-target area, including:

[0054] Determine a feature area according to the first image data of the sub-target area, perform image enhancement and verification on the feature area in the second image data of the sub-target area, and transmit the verified second image data to the supervision end;

[0055] Alternatively, global image enhancement is performed on the second image data of the sub-target area and the image is transmitted to the monitoring end.

[0056] Specifically, there is no limitation on the specific structure of the mobile image acquisition unit, which only needs to have the function of acquiring images of the construction site, for example, a drone equipped with an image acquisition device, which will not be elaborated here.

[0057] Specifically, the mobile image acquisition unit moves along a predetermined route and acquires image data at a fixed acquisition point. It can be understood that since the first image data is used for reference, the shooting angle and shooting ratio of the image should be the same when acquiring the first image data and the second image data, which will not be elaborated here.

[0058] Specifically, there is no limitation on the division of the sub-target areas. It is understandable that the sizes of the divided sub-target areas should be the same.

[0059] Specifically, there is no limitation on the method of determining the contour of an object in the image data. For example, an image segmentation algorithm may be used to identify the contour of the object. Of course, other methods may also be used, which will not be described in detail.

[0060] Specifically, there is no limitation on the method of determining the texture density in the contour of the object in the image data. For example, the texture features in the image are enhanced by filtering technology to determine the texture density. Of course, other methods can also be used, which will not be described in detail.

[0061] Specifically, in step S2, the object salient feature characterization parameter is calculated based on the object salient feature according to formula (1), including:

[0062]

[0063] In formula (1), R represents the parameter representing the object's prominent features, fe(x) represents the feature prominent function, N represents the number of object contours, N0 represents the number threshold, M represents the texture density in the object contour, M0 represents the texture density threshold, α represents the number weight coefficient of the object contour, β represents the texture density weight coefficient in the object contour, and e represents a natural constant.

[0064] Specifically, in this example, the number threshold N0 of object contours and the texture density threshold M0 in object contours are obtained in advance, and several image data for different sub-target areas of the construction site are obtained to determine the number of object contours and the texture density data in the object contours in each sub-target area. The mean number of object contours ∆N0 and the mean texture density ∆M0 in the object contours are solved, and N0=r1×∆N0 and M0=r2×∆M0 are set, where r1 represents the first precision coefficient and r2 represents the second precision coefficient, where 0.65<r1<0.85 and 0.35<r2<0.65, α is 0.55 and β is 0.45.

[0065] The present invention performs unified image acquisition on each target area of ​​the construction site in advance without meeting weak light conditions to obtain first image data as a reference. The first image data is collected without meeting weak light conditions, is relatively clear, and has obvious features. In actual situations, when performing image acquisition on a large-scale construction site, it is important to monitor feature areas, such as material accumulation areas, construction areas, etc. However, for large-scale construction areas, there may also be a large number of featureless areas, such as vacant land, which is reflected in the fact that there are no obvious features and no valuable information in the image data, and the image data is single, which is not easy to perform feature extraction and analysis, and the analysis value is low. Therefore, the present invention considers pre-determining the object salient features of each target area through the first image data, and correspondingly calculating the object salient feature characterization parameters. Since the object salient features include the number of object contours and the texture density in the object contours, the calculated object salient feature characterization parameters can characterize the value information contained in different target areas, and provide support for the subsequent adaptive analysis of the second image data. Furthermore, the massive image data is processed in a targeted manner, and the computing power consumption brought by analyzing the image data is reduced while ensuring reliability.

[0066] Specifically, in step S2, determining the feature highlight category of the sub-target area includes:

[0067] If the object salient feature representation parameter is greater than or equal to the object salient feature representation parameter threshold, the feature salient feature of the sub-target area is determined to be a strong feature salient category;

[0068] If the object salient feature representation parameter is less than the object salient feature representation parameter threshold, the feature salient feature of the sub-target area is determined to be a weak feature salient category.

[0069] The object salient feature characterization parameter threshold R0 is selected in the interval [1.32, 1.41].

[0070] The present invention divides the feature highlighting categories of the sub-target area based on the first image data, and then, different collection accuracies are used when acquiring the second image data subsequently. For the strong feature highlighting area, since the feature area has more features, it is collected with higher accuracy, which is convenient for the subsequent global image enhancement of the collected second highlighting data for the strong feature highlighting area. For the weak feature highlighting area, it is considered that there may be subdivided areas without obvious features and valuable information. Therefore, for the image data collected for the weak feature highlighting area, the feature area with features is determined, and only the feature area is enhanced to avoid feature omission, thereby reducing the computing power consumption brought by analyzing the image data while ensuring reliability.

[0071] Specifically, in step S3, the image acquisition accuracy for the sub-target area is determined according to the feature highlighting category of the sub-target area, including:

[0072] If the feature highlighting category of the sub-target area is a weak feature highlighting category, it is determined that the reference image acquisition accuracy is used when acquiring the sub-target area;

[0073] If the feature highlighting category of the sub-target area is a strong feature highlighting category, the base image acquisition accuracy is increased when acquiring the sub-target area.

[0074] It is understandable that adjusting the image acquisition accuracy can be achieved by adjusting the image resolution. For sub-target areas of image categories with strong features, the image resolution can be appropriately increased to better highlight the features.

[0075] It is understandable that for monitoring of large areas, in order to reduce the amount of data, the baseline image acquisition accuracy can be 576i or 480i when low-light conditions are not met, and 720p or 1080p can be used for the increased image acquisition accuracy.

[0076] Specifically, in step S4, the second image data of the sub-target area is processed according to the feature highlighting category of the sub-target area, including:

[0077] If the feature highlighting of the sub-target area is a weak feature highlighting category, determining the feature area according to the first image data of the sub-target area, performing image enhancement and verification on the feature area in the second image data of the sub-target area, and transmitting the verified second image data to the supervision end;

[0078] If the feature highlighting of the sub-target area is a strong feature highlighting category, global image enhancement is performed on the second image data of the sub-target area and transmitted to the supervision end.

[0079] It is understandable that the monitoring end may be a display device with the function of displaying images.

[0080] Specifically, in step S4, the process of determining the feature area according to the first image data of the sub-target area includes:

[0081] Dividing the first image data into a plurality of image regions and setting sequence numbers;

[0082] Determine an image region that meets a feature screening condition and an adjacent image region of the image region as a feature region;

[0083] The feature screening condition is that there is an object contour in the image area and the area of ​​the object contour must be greater than a predetermined area threshold.

[0084] It can be understood that in order to eliminate the influence of smaller features, the area threshold is selected within the interval [1%Se, 5%Se], and Se is set to represent the total area of ​​the second image data.

[0085] Specifically, in step S4, the process of performing image enhancement on the feature area in the second image data of the sub-target area includes:

[0086] Dividing the second image data into a plurality of image regions and setting sequence numbers;

[0087] The image region to be enhanced in the second image data is determined based on the sequence number of each image region in the determined feature region.

[0088] It can be understood that the division method of the first image data and the second image data is the same, and the serial number setting method is also the same to reflect the corresponding relationship.

[0089] Specifically, in step S4, the verification process includes:

[0090] determining a sub-target area corresponding to the second image data;

[0091] Extracting first image data corresponding to the sub-target area;

[0092] Determine the degree of overlap of the object contour of the first image data and the second image data;

[0093] If the object contour coincidence is greater than a predetermined coincidence threshold, it is determined that the second image data meets the verification standard.

[0094] Specifically, there is no limitation on the specific method for determining the overlap of object contours. For example, the overlap of object contours can be measured by Euclidean distance. The Euclidean distances of corresponding contour points in the image are calculated, and the corresponding Euclidean distances of all contour points are averaged as the overlap of object contours. Of course, other methods can also be used, which will not be described in detail.

[0095] The predetermined coincidence threshold G0 is predetermined, wherein the second image data for a single sub-target area is obtained as a sample for global image enhancement, the contours of each object in the sample are determined, and the object contour coincidence mean Ge of each object contour is solved, and it is set that G0=g×Ge, g represents the contour deviation coefficient, 0.85<g<0.95.

[0096] Specifically, in step S1, the mobile image acquisition unit moves along a predetermined route and acquires image data at a fixed acquisition point.

[0097] It is understandable that the scheduled route can be selected based on the division of the sub-target areas, which will not be described in detail.

[0098] Specifically, the low-light condition is that the current ambient light intensity is less than a predetermined light intensity threshold.

[0099] In order to reflect the ambient light intensity at night, the light intensity threshold is selected in the interval [0.15lx, 0.25Ix].

[0100] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A low-light image enhancement method for a construction site, characterized in that: include: Step S1, controlling the mobile image acquisition unit to acquire a target area under a weak light condition to obtain a plurality of first image data covering the target area; Step S2, extracting first image data for different sub-target areas, identifying object salient features in the first image data, and calculating object salient feature characterization parameters based on the object salient features to determine feature salient categories of the sub-target areas, wherein the object salient features include the number of object contours and the texture density in the object contours; Step S3, determining the image acquisition accuracy for the sub-target area according to the feature highlighting category of the sub-target area, acquiring the target area with the corresponding image acquisition accuracy under weak light conditions, and obtaining a plurality of second image data covering the target area; Step S4, processing the second image data of the sub-target area according to the feature highlighting category of the sub-target area, including: Determine a feature area according to the first image data of the sub-target area, perform image enhancement and verification on the feature area in the second image data of the sub-target area, and transmit the verified second image data to the supervision end; Or, performing global image enhancement on the second image data of the sub-target area and transmitting the same to the supervision end; In step S2, the object salient feature characterization parameter is calculated based on the object salient feature according to formula (1), including: In formula (1), R represents the parameter representing the object's prominent features, fe(x) represents the feature prominent function, N represents the number of object contours, N0 represents the number threshold, M represents the texture density in the object contour, M0 represents the texture density threshold, α represents the number weight coefficient of the object contour, β represents the texture density weight coefficient in the object contour, and e represents a natural constant; The number threshold N0 of object contours and the texture density threshold M0 in object contours are obtained in advance. Several image data for different sub-target areas of the construction site are obtained, and the number of object contours and the texture density data in the object contours in each sub-target area are determined. The mean number of object contours ∆N0 and the mean texture density ∆M0 in the object contours are solved, and N0=r1×∆N0 and M0=r2×∆M0 are set, where r1 represents the first precision coefficient and r2 represents the second precision coefficient.

2. The method for enhancing weak-light images at construction sites according to claim 1, characterized in that: In the step S2, determining the feature highlight category of the sub-target area includes: If the object salient feature representation parameter is greater than or equal to the object salient feature representation parameter threshold, the feature salient feature of the sub-target area is determined to be a strong feature salient category; If the object salient feature representation parameter is less than the object salient feature representation parameter threshold, the feature salient feature of the sub-target area is determined to be a weak feature salient category.

3. The method for enhancing weak-light images at construction sites according to claim 1, characterized in that: In step S3, the image acquisition accuracy for the sub-target area is determined according to the feature highlighting category of the sub-target area. include, If the feature highlighting category of the sub-target area is a weak feature highlighting category, it is determined that the reference image acquisition accuracy is used when acquiring the sub-target area; If the feature highlighting category of the sub-target area is a strong feature highlighting category, the base image acquisition accuracy is increased when acquiring the sub-target area.

4. The method for enhancing weak-light images at construction sites according to claim 1, characterized in that: In step S4, the second image data of the sub-target area is processed according to the feature highlighting category of the sub-target area, including: If the feature highlighting of the sub-target area is a weak feature highlighting category, determining the feature area according to the first image data of the sub-target area, performing image enhancement and verification on the feature area in the second image data of the sub-target area, and transmitting the verified second image data to the supervision end; If the feature highlighting of the sub-target area is a strong feature highlighting category, global image enhancement is performed on the second image data of the sub-target area and transmitted to the supervision end.

5. The method for enhancing weak-light images at construction sites according to claim 1, characterized in that: In step S4, the process of determining the feature area according to the first image data of the sub-target area includes: Dividing the first image data into a plurality of image regions and setting sequence numbers; Determine an image region that meets a feature screening condition and an adjacent image region of the image region as a feature region; The feature screening condition is that there is an object contour in the image area and the area of ​​the object contour must be greater than a predetermined area threshold.

6. The weak-light image enhancement method for a construction site according to claim 1, characterized in that: In step S4, the process of performing image enhancement on the feature area in the second image data of the sub-target area includes: Dividing the second image data into a plurality of image regions and setting sequence numbers; The image region to be enhanced in the second image data is determined based on the sequence number of each image region in the determined feature region.

7. The method for enhancing weak-light images at construction sites according to claim 1, characterized in that: In step S4, the verification process includes: determining a sub-target area corresponding to the second image data; Extracting first image data corresponding to the sub-target area; Determine the degree of overlap of the object contour of the first image data and the second image data; If the object contour coincidence is greater than a predetermined coincidence threshold, it is determined that the second image data meets the verification standard.

8. The method for enhancing weak-light images at construction sites according to claim 1, characterized in that: In step S1, the mobile image acquisition unit moves along a predetermined route and acquires image data at a fixed acquisition point.

9. The method for enhancing weak-light images at construction sites according to claim 1, characterized in that: The low light condition is that the current ambient light intensity is less than a predetermined light intensity threshold.

Images