An environmental sensing-based surgical lighting control system and method

Abstract

This invention discloses a surgical lighting control system and method based on environmental sensing, belonging to the field of lighting control technology. The invention acquires images of the surgical site, obtains target coordinates and boundary coordinate point sets through feature matching; based on the target coordinates, a mapping relationship library is matched to determine the set of LEDs covering the coordinate area; the current actual brightness value of each LED is calculated; the shadow area formed on the target coordinates after the doctor blocks the light is extracted, and the shadow area is calculated; the set of blocked LEDs and the set of unblocked LEDs are identified; for the set of blocked LEDs, the brightness deviation between each LED and the height and angle deviation of each LED group are calculated; the brightness reduction value and total reduction amount of the blocked LEDs are calculated, and the brightness compensation parameters of the unblocked LEDs are determined; a correlation function between the shadow area and the LED group posture parameters is generated to obtain the posture adjustment amount; brightness control signals and posture control signals are generated to control the surgical lights.

Description

Technical Field

[0001] This invention relates to the field of lighting control technology, specifically to a surgical lighting control system and method based on environmental sensing. Background Technology

[0002] In surgical procedures, surgical lights, as core auxiliary equipment, directly determine the clarity of the surgical field, thereby affecting the precision, safety, and patient prognosis of the surgical operation. As surgical procedures become more minimally invasive and refined, clinical demands for surgical lighting have increased: precise focused illumination of the surgical site is required, avoiding shadows created by the surgeon's body or instruments blocking the light during the procedure, while ensuring stable and uniform lighting brightness, and dynamically adapting to the surgical scenario.

[0003] Current technologies largely rely on manual adjustment after shadows are detected, or indirect identification of occlusion using a single sensor. They cannot accurately identify the occluded area and the corresponding obscured LEDs in real time. Control methods often involve increasing overall brightness or roughly adjusting the lamp assembly's posture, failing to eliminate shadows at their root. Existing technologies cannot distinguish between obscured and unobscured LEDs, often adjusting the brightness of all LEDs uniformly. This prevents the proper release of redundant brightness from obscured LEDs, and fails to provide precise compensation to unobscured LEDs, resulting in fluctuations in overall brightness in the surgical area due to occlusion. Furthermore, in existing technologies, lamp assembly posture adjustment and brightness adjustment are mostly controlled independently, lacking a correlation model between shadow area and posture parameters. This prevents the optimization of lighting effects through the synergistic effect of posture adjustment and brightness compensation. Summary of the Invention

[0004] The purpose of this invention is to provide an environmental sensing-based surgical lighting control system and method to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] In a first aspect, this application provides a surgical lighting control method based on environmental sensing, comprising the following steps:

[0007] Images of the surgical site are acquired, and the target coordinates and boundary coordinate point set are obtained through feature matching. Based on the target coordinates, a matching mapping database is used to determine the set of LED beads covering the coordinate region jointly defined by the target coordinates and boundary coordinate point set.

[0008] Collect occlusion signals and brightness signals of the surgical lighting area, and calculate the current actual brightness value of each LED; based on the target coordinates and boundary coordinate point set, extract the shadow area formed on the target coordinates after the doctor blocks the light, and calculate the shadow area;

[0009] Based on the current actual brightness value and shadow area of ​​each LED, identify the set of LEDs that are blocked and the set of LEDs that are not blocked; for the set of LEDs that are blocked, calculate the brightness deviation between each LED and the height and angle deviation of each LED group;

[0010] Based on the brightness deviation between each LED, the brightness reduction ratio of the blocked LED is determined, the required brightness reduction value and total reduction amount of the blocked LED are calculated, and the brightness compensation parameters of the unblocked LED are determined. Based on the height and angle deviations of each LED group, a correlation function between the shadow area and the LED group attitude parameters is generated to obtain the attitude adjustment amount. Based on the required brightness reduction value of the blocked LED and the brightness compensation parameters of the unblocked LED, a brightness control signal is generated, and based on the attitude adjustment amount, an attitude control signal is generated to control the surgical light.

[0011] In conjunction with the first aspect, in a first embodiment of the first aspect of this application, the step of acquiring an image of the surgical site and obtaining a set of target coordinates and boundary coordinate points through feature matching includes:

[0012] The image acquisition camera array is started, and the image recognition parameters are initialized. The camera array continuously acquires images of the surgical operation area according to the initialized frame rate, obtaining a raw image sequence containing the surgical site and the surrounding environment. Image preprocessing is performed on the acquired raw image sequence, including grayscale conversion, Gaussian filtering for noise reduction, and image enhancement, and the preprocessed image set is output.

[0013] Based on the preset lesion feature template, a feature matching algorithm is executed on the preprocessed image set. By calculating the similarity between each region in the image and the feature template, regions with similarity greater than the feature matching threshold are selected as lesion regions, and the pixel coordinate point set of the lesion region is obtained. This set is then converted into a three-dimensional spatial dimension to obtain spatial coordinate data. Combined with the position markers of each camera, the spatial coordinate data collected by multiple cameras are fused to output the target coordinates and boundary coordinate point set.

[0014] In conjunction with the first aspect, in the second embodiment of the first aspect of this application, the step of determining the set of LED beads covering the coordinate region jointly defined by the target coordinates and the boundary coordinate point set based on the target coordinates and matching the mapping relationship library includes:

[0015] The system calls a mapping library containing 3D coordinate partitioning rules for the surgical lighting area, illumination characteristic parameters of individual LEDs, and preset associations between coordinate sub-partitions and LEDs. The target coordinates and boundary coordinate point sets are substituted one by one into the 3D coordinate partitioning rules. When the X, Y, and Z axis values ​​of a coordinate point all fall within the coordinate threshold range of a certain sub-partition, the coordinate point is determined to belong to that sub-partition. All sub-partitions to which coordinate points belong are summarized to form a partition set. The preset associations between coordinate sub-partitions and LEDs are called to extract all LEDs corresponding to each sub-partition in the partition set, forming a candidate LED list L.

[0016] Obtain the actual irradiation distance between the surgical light and the affected area, and call the irradiation characteristic parameters of each lamp bead in L; for each lamp bead in L, calculate its effective irradiation range at the actual irradiation distance based on the relationship between the lamp bead's irradiation angle, irradiation distance and coverage area, and determine whether the effective irradiation range intersects with the target coordinate area; retain the lamp beads that are determined to have intersection, and remove the lamp beads that do not intersect, forming a set of lamp beads that covers the coordinate area jointly defined by the target coordinates and the boundary coordinate point set.

[0017] In conjunction with the first aspect, in the third embodiment of the first aspect of this application, the step of collecting the occlusion signal of the surgical lighting area and the brightness signal of each area, and calculating the current actual brightness value of each LED, includes:

[0018] The photoelectric switch sensor array scans the entire surgical illumination area at a preset frequency to capture occlusion signals generated by obstructions; it collects brightness data of the illumination area according to preset zones to obtain brightness signals of each zone; it performs noise reduction and filtering on the occlusion signals and brightness signals, removes interference signals below a preset intensity threshold, and inputs them into an A / D converter. According to the initial sampling accuracy and conversion rate, the analog signals are converted into digital signals to obtain digital occlusion signals and digital brightness signals.

[0019] A convolutional neural network is used to train a model of the mapping relationship between LED current and brightness. Combined with the digital brightness signal, the current power supply current signal of the corresponding area's lamp group is inferred. Based on the correlation between the lamp group and the LED, the current power supply current signal is allocated to individual LEDs, and the actual brightness value of each LED is calculated by substituting it into the mapping relationship model.

[0020] In conjunction with the first aspect, in the fourth embodiment of the first aspect of this application, the step of extracting the shadow area formed on the target coordinates after the doctor blocks the light, and calculating the shadow area based on the target coordinates and the boundary coordinate point set, includes:

[0021] Based on the target coordinates and boundary coordinate point set, the spatial region corresponding to the target coordinates is defined in the image to generate the target region ROI. The ROI represents the region of interest, and the background region in the image that is not related to the surgical site is excluded. The target region ROI is then cropped and scaled to unify the image size specifications.

[0022] A grayscale feature analysis algorithm is used to perform a full-domain scan of the pixel grayscale values ​​within the target region ROI to distinguish low-brightness regions with grayscale values ​​below a preset threshold. Combining the spatial correspondence between the shadow region and the location of the occluder, a region connectivity analysis algorithm is used to filter out low-brightness regions with continuous boundaries and whose locations match the occluder regions detected by the photoelectric switch sensor, thus eliminating isolated low-brightness noise points. An edge detection algorithm is used to locate the complete boundary of the filtered shadow region, forming a closed shadow region outline.

[0023] Based on the calibration parameters of the image acquisition camera, the actual spatial area corresponding to a single pixel in the image is determined. A regional pixel statistics algorithm is used to count all pixels within the extracted shadow region contour to obtain the total number of pixels corresponding to the shadow region. Combining the actual spatial area corresponding to a single pixel, the total number of pixels in the shadow region is converted into the actual spatial area through multiplication. At the same time, an error correction algorithm is used to correct the conversion result to obtain the shadow area.

[0024] In conjunction with the first aspect, in the fifth embodiment of the first aspect of this application, the step of identifying the set of blocked LEDs and the set of unblocked LEDs based on the current actual brightness value and shadow area of ​​each LED includes:

[0025] Extract the preset reference brightness value of each LED and set a brightness deviation threshold; iterate through the current actual brightness value of all LEDs and calculate the deviation value between the actual brightness of each LED and the reference brightness. When the deviation value exceeds the preset threshold, the LED is included in the low brightness LED candidate set; based on the shadow area, extract the complete spatial coordinate range of the shadow area; iterate through each LED in the low brightness LED candidate set and call its corresponding illumination area coordinates; through the spatial range intersection judgment logic, determine whether the illumination range of the LED overlaps with the spatial range of the shadow area; retain LEDs whose illumination range overlaps with the shadow area, remove LEDs that do not overlap, form the set of blocked LEDs, and include the remaining LEDs in the set of unblocked LEDs.

[0026] In conjunction with the first aspect, in the sixth embodiment of the first aspect of this application, the step of calculating the brightness deviation between each LED and the height and angle deviation of each LED group for the set of blocked LEDs includes:

[0027] For each blocked LED in the same blocked LED group, the average current actual brightness of all blocked LEDs in the group is used as the group's reference brightness. When cross-group comparison is required, an additional average brightness of the blocked LEDs is set as the global reference brightness. Each blocked LED in the same blocked LED group is iterated through, and the difference between its current actual brightness and the group's reference brightness is calculated to obtain the group's brightness deviation for a single LED. All blocked LEDs in the group are paired, and the brightness difference between each pair is calculated to form a brightness deviation matrix between LEDs in the group. The positive and negative attributes of each brightness deviation are labeled, with positive values ​​indicating that the LED's brightness is higher than the reference or comparison LED, and negative values ​​indicating that the LED's brightness is lower than the reference or comparison LED.

[0028] For each lamp group to which the obscured LED belongs, extract its preset reference height and current actual height, calculate the difference between the two, and obtain the height deviation of the lamp group; clarify the meaning of positive and negative deviations, a positive value indicates that the current height is lower than the reference height, and a negative value indicates that the current height is higher than the reference height; extract the preset reference rotation angle and current actual rotation angle of each lamp group to which the obscured LED belongs, calculate the difference between the two, and obtain the angle deviation of the lamp group; clarify the meaning of positive and negative deviations, a positive value indicates that the current rotation angle deviates from the reference angle counterclockwise, and a negative value indicates that it deviates from the reference angle clockwise.

[0029] In conjunction with the first aspect, in the seventh embodiment of the first aspect of this application, the step of determining the brightness reduction ratio of the blocked LED based on the brightness deviation between each LED, calculating the brightness reduction value and total reduction amount of the blocked LED, and determining the brightness compensation parameters of the unblocked LEDs includes:

[0030] A deviation grading algorithm is used to divide the brightness deviation between the blocked LED and other LEDs in the same group into several consecutive levels. Each level is matched with a preset basic brightness reduction ratio range. The current actual brightness value of each blocked LED is used as the base, and multiplied by its corresponding final brightness reduction ratio to obtain the basic brightness reduction value of each LED. The brightness reduction values ​​of all blocked LEDs are summed to obtain the total brightness reduction of the set of blocked LEDs.

[0031] The difference between the overall target illuminance of the surgical light and the current overall actual illuminance is calculated. An overlay operation logic is used to add the basic illuminance gap to the total brightness reduction of the blocked LEDs, yielding the total compensation brightness requirement for the unblocked LEDs. A balanced allocation algorithm combined with capability adaptation is employed, considering the current actual brightness and maximum brightness carrying capacity of the unblocked LEDs, to distribute the total compensation brightness requirement to each unblocked LED. Finally, the current actual brightness value of the unblocked LEDs is multiplied by their corresponding compensation ratio to obtain the basic increase in brightness required for each LED.

[0032] In conjunction with the first aspect, in the eighth embodiment of the first aspect of this application, the step of generating a correlation function between the shadow area and the attitude parameters of the light group based on the height and angle deviations of each light group to obtain the attitude adjustment amount includes:

[0033] The topology node types are determined. Input nodes include shadow area, height deviation, and angle deviation. Output nodes include target height and rotation angle of the light group. Intermediate nodes include target coordinates and illumination range of the light group. Edges represent the correlation strength between nodes, and the basic framework of the topology structure is built. The correlation degree between each node is calculated through a correlation analysis algorithm. Nodes with a correlation degree higher than a preset threshold are connected with edges to form a topology network that includes the influence of posture deviation. A topology optimization algorithm is used to remove redundant edges with too low correlation strength and merge duplicate nodes.

[0034] A topology analysis algorithm is used to identify the dependency paths of each node in the topology network and determine the transmission logic. Specifically, using the shadow area as the core response variable and considering the effects of height and angle deviations, the key transmission path from the input node to the output node is derived, and the association rules and quantification parameters on the path are extracted. Based on the transmission paths and association strengths of the topology nodes, and combined with a data fitting algorithm, the relationship between the shadow area and attitude deviation and the target attitude parameters of the lamp group is transformed into a function, forming a preliminary association function. Historical valid data samples are substituted into the preliminary function, and an error analysis algorithm is used to calculate the deviation between the target attitude parameters output by the function and the actual data. When the deviation exceeds the allowable threshold, the process returns to the topology network optimization step to adjust the parameters and regenerate the function until the deviation meets the requirements, thus determining the association function between the shadow area and the attitude parameters of the lamp group.

[0035] Substitute the real-time shadow area, real-time height deviation, and real-time angle deviation data of the current surgery into the function, and obtain the target posture parameters of the lamp group required to reduce the shadow area to the preset allowable range through function calculation; calculate the difference between the current posture parameters of the lamp group and the target posture parameters output by the associated function to obtain the height adjustment amount and the rotation angle adjustment amount, and mark the direction of the adjustment amount.

[0036] Secondly, this application provides an environmental sensing-based surgical lighting control system, comprising:

[0037] Target localization and LED bead set matching module: including: target coordinate extraction unit acquires images of the surgical site and obtains target coordinates and boundary coordinate point set through feature matching; LED bead set matching unit determines the LED bead set covering the coordinate region jointly defined by the target coordinates and boundary coordinate point set based on the target coordinates and matching mapping relationship library.

[0038] The shadow area calculation module includes: a lamp brightness inverse calculation unit that collects the occlusion signal and brightness signal of each area in the surgical lighting area and inversely calculates the current actual brightness value of each lamp; and a shadow area calculation unit that extracts the shadow area formed on the target coordinates after the doctor blocks the light based on the target coordinates and boundary coordinate point set, and calculates the shadow area.

[0039] Deviation calculation module: includes: a shading and unshading set identification unit that identifies the shading set and the unshading set of lamp beads based on the current actual brightness value and shadow area of ​​each lamp bead; and a deviation calculation unit that calculates the brightness deviation between each lamp bead and the height and angle deviation of each lamp group for the shading set of lamp beads.

[0040] The control execution module includes: a brightness adjustment parameter calculation unit that determines the brightness reduction ratio of the blocked LEDs based on the brightness deviation between each LED, calculates the required brightness reduction value and total reduction amount for the blocked LEDs, and determines the brightness compensation parameters for the unblocked LEDs; a posture adjustment parameter calculation unit that generates a correlation function between the shadow area and the posture parameters of each light group based on the height and angle deviations of each light group, and obtains the posture adjustment amount; and a control execution unit that generates a brightness control signal based on the required brightness reduction value of the blocked LEDs and the brightness compensation parameters for the unblocked LEDs, and generates a posture control signal based on the posture adjustment amount to control the surgical lights.

[0041] Compared with the prior art, the beneficial effects of the present invention are:

[0042] 1. This invention can accurately distinguish between shaded and unshaded sets of LED beads by collecting shading signals, brightness signals and shadow area data. It can also achieve dynamic control by combining the brightness deviation between LED beads and the posture deviation of the lamp group, thus transforming the passive response to shadows into active prevention and precise elimination.

[0043] 2. This invention sets a personalized brightness reduction ratio for the blocked LED beads to release redundant brightness; at the same time, it accurately allocates compensation parameters for the unblocked LED beads, and achieves coordinated regulation of current reduction and current increase through circuit control to ensure that the overall brightness of the surgical area is stable at the target value and avoids brightness fluctuations caused by blockage.

[0044] 3. This invention establishes a correlation function between shadow area and posture parameters based on the posture deviation of the lamp group. Through the coordinated output of posture adjustment amount and brightness control signal, it realizes dual control of posture optimization and brightness compensation, which can adapt to the dynamic occlusion requirements of different surgical operation scenarios. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the steps of a surgical lighting control method based on environmental sensing according to the present invention;

[0046] Figure 2This is a system structure diagram of a surgical lighting control system based on environmental sensing according to the present invention. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Example: Figures 1-2 As shown, the present invention provides a technical solution:

[0049] like Figure 1 As shown, this application provides a surgical lighting control method based on environmental sensing, including the following steps:

[0050] Step S100: Acquire an image of the surgical site, obtain the target coordinates and boundary coordinate point set through feature matching; based on the target coordinates, match the mapping relationship library to determine the set of LED beads covering the coordinate region jointly defined by the target coordinates and boundary coordinate point set;

[0051] Specifically, the image acquisition camera array is started, the image recognition parameters are initialized, and the camera array continuously acquires images of the surgical operation area according to the initialized frame rate to obtain the original image sequence containing the surgical site and the surrounding environment; image preprocessing is performed on the acquired original image sequence, including grayscale conversion, Gaussian filtering for noise reduction and image enhancement, and the preprocessed image set is output;

[0052] Based on the preset lesion feature template, a feature matching algorithm is executed on the preprocessed image set. By calculating the similarity between each region in the image and the feature template, regions with similarity greater than the feature matching threshold are selected as lesion regions, and the pixel coordinate point set of the lesion region is obtained. This set is then converted into a three-dimensional spatial dimension to obtain spatial coordinate data. Combined with the position markers of each camera, the spatial coordinate data collected by multiple cameras are fused to output the target coordinates and boundary coordinate point set.

[0053] Furthermore, the mapping relationship library is invoked, which includes the three-dimensional coordinate partitioning rules for the surgical lighting area, the illumination characteristic parameters of a single LED, and the preset association relationships between coordinate sub-partitions and LEDs. The target coordinates and boundary coordinate point sets are substituted one by one into the three-dimensional coordinate partitioning rules. When the X, Y, and Z axis values ​​of a coordinate point all fall within the coordinate threshold range of a certain sub-partition, the coordinate point is determined to belong to that sub-partition. The sub-partitions to which all coordinate points belong are summarized to form a partition set. The preset association relationships between coordinate sub-partitions and LEDs are invoked to extract all LEDs corresponding to each sub-partition in the partition set, forming a candidate set L of LED list.

[0054] Obtain the actual irradiation distance between the surgical light and the affected area, and call the irradiation characteristic parameters of each lamp bead in L; for each lamp bead in L, calculate its effective irradiation range at the actual irradiation distance based on the relationship between the lamp bead's irradiation angle, irradiation distance and coverage area, and determine whether the effective irradiation range intersects with the target coordinate area; retain the lamp beads that are determined to have intersection, and remove the lamp beads that do not intersect, forming a set of lamp beads that covers the coordinate area jointly defined by the target coordinates and the boundary coordinate point set.

[0055] In one specific embodiment, based on an orthopedic joint replacement surgery scenario, the surgical light employs an array of 80 LEDs (maximum illuminance of 160,000 LUX) and is equipped with a data acquisition array consisting of three sets of high-definition industrial cameras, installed 1.4 meters directly above the operating table. The standard illumination distance is 1 meter, but in this experiment, the actual illumination distance was adjusted to 1.2 meters (>1 meter standard distance). After startup, the image recognition parameters were initialized, the camera acquisition frame rate was set to 30fps, and continuous acquisition of images of the knee joint and surrounding area was performed for 10 seconds, obtaining a raw image sequence of 300 frames with a resolution of 1920×1080 pixels.

[0056] Preprocessing is performed on the original image sequence: first, it is converted into a single-channel grayscale image with a grayscale value of 0-255, then denoising is performed using a 5×5 Gaussian filter kernel (standard deviation 1.2) to remove instrument reflection interference, and finally image contrast is enhanced by histogram equalization (by 40%), outputting a set of 300 preprocessed images.

[0057] A pre-defined knee joint feature template was used, and the SIFT algorithm was employed for feature matching. A matching threshold of 0.85 was set to filter out affected areas with similarity scores, corresponding to pixel coordinates (450-880, 360-730). The data was converted to 3D spatial coordinates using camera calibration parameters: X-axis 580-850mm, Y-axis 500-770mm, Z-axis 700-760mm. Data from three cameras (located at 550mm, 750mm, and 950mm on the X-axis, respectively) was fused to output target coordinates (715, 635, 730)mm, with boundary coordinate point sets of (580, 500, 700)mm, (850, 500, 700)mm, (850, 770, 760)mm, and (580, 770, 760)mm.

[0058] The mapping relationship library is invoked, and its 3D coordinate partitioning rules divide the Z-axis range of 1.0-1.5 meters into 10 sub-partitions (adapting to an 80-LED layout). Each sub-partition has an X and Y axis span of 65mm, and the association between the sub-partitions and the 80 LEDs is pre-stored. The coordinate points are then used for judgment; if all points fall within 3 sub-partitions, the corresponding LEDs are extracted to form a candidate set L (containing 25 LEDs, numbered 20-32 and 45-52).

[0059] The laser ranging module confirmed the actual illumination distance was 1.2 meters. It then retrieved the illumination characteristic parameters of the candidate set of LED beads (single bead illumination angle 14°, coverage area 0.018㎡ at 1 meter distance, coverage area increases by 0.004㎡ for every 100mm increase in distance). The effective illumination range of each bead was calculated, and it was determined that the illumination ranges of 15 beads (numbered 21-28, 46-50) overlapped with the target area. Ten beads without overlap were removed, forming the final set of covering LED beads.

[0060] Step S200: Collect the occlusion signal and brightness signal of each area in the surgical lighting area, and calculate the current actual brightness value of each lamp bead; based on the target coordinates and boundary coordinate point set, extract the shadow area formed on the target coordinates after the doctor blocks the light, and calculate the shadow area;

[0061] Specifically, the photoelectric switch sensor array scans the entire surgical illumination area at a preset frequency to capture occlusion signals generated by obstructions; it collects brightness data of the illumination area according to preset zones to obtain brightness signals of each zone; it performs noise reduction and filtering on the occlusion signals and brightness signals, removes interference signals below a preset intensity threshold, and inputs them into an A / D converter. Based on the initial sampling accuracy and conversion rate, the analog signals are converted into digital signals to obtain digital occlusion signals and digital brightness signals.

[0062] A convolutional neural network is used to train a model of the mapping relationship between LED current and brightness. Combined with the digital brightness signal, the current power supply current signal of the corresponding area's lamp group is inferred. Based on the correlation between the lamp group and the LED, the current power supply current signal is allocated to individual LEDs, and the actual brightness value of each LED is calculated by substituting it into the mapping relationship model.

[0063] Furthermore, based on the target coordinates and boundary coordinate point set, the spatial region corresponding to the target coordinates is defined in the image to generate the target region ROI. The ROI represents the region of interest, excluding background areas in the image that are unrelated to the surgical site. The target region ROI is then cropped and scaled to unify the image size specifications.

[0064] A grayscale feature analysis algorithm is used to perform a full-domain scan of the pixel grayscale values ​​within the target region ROI to distinguish low-brightness regions with grayscale values ​​below a preset threshold. Combining the spatial correspondence between the shadow region and the location of the occluder, a region connectivity analysis algorithm is used to filter out low-brightness regions with continuous boundaries and whose locations match the occluder regions detected by the photoelectric switch sensor, thus eliminating isolated low-brightness noise points. An edge detection algorithm is used to locate the complete boundary of the filtered shadow region, forming a closed shadow region outline.

[0065] Based on the calibration parameters of the image acquisition camera, the actual spatial area corresponding to a single pixel in the image is determined. A regional pixel statistics algorithm is used to count all pixels within the extracted shadow region contour to obtain the total number of pixels corresponding to the shadow region. Combining the actual spatial area corresponding to a single pixel, the total number of pixels in the shadow region is converted into the actual spatial area through multiplication. At the same time, an error correction algorithm is used to correct the conversion result to obtain the shadow area.

[0066] In one specific embodiment, the brightness acquisition area is divided into 10 sub-zones according to S100. Occlusion signals and brightness signals are captured simultaneously. The occlusion signal strength range is 0-5V, with a preset strength threshold of 0.3V. Four sets of environmental interference signals are removed. The brightness signal range is 500-150000LUX, with no invalid signals. The valid signals are input to a 16-bit A / D converter, with sampling accuracy set to 0.01V (occlusion signal) and 10LUX (brightness signal), and a conversion rate of 100kHz. Finally, the occlusion digital signal (0.4-3.5V) and brightness digital signal (800-148000LUX) for the 10 zones are obtained.

[0067] A convolutional neural network was used to train the LED current-brightness mapping model. The training samples consisted of 1200 sets of current (0.1-0.6A) and brightness data for 80 LEDs, achieving a fitting accuracy of 98.8%. The brightness digital signals of 10 zones were input into the model to inversely deduce the current supply current (0.13-0.58A) of the corresponding lamp group. The current was then allocated to individual LEDs according to the "lamp group-LED" relationship, and the actual brightness of each LED was calculated: the brightness range of the 15 covering LEDs (21-28, 46-50) determined by S100 was 135000-148000 LUX, and the brightness range of the remaining 65 LEDs was 800-95000 LUX, all of which did not exceed the upper limit of 160,000 LUX.

[0068] Based on the target coordinates and boundary coordinate point set output by S100, the ROI region is defined in the preprocessed image, corresponding to the pixel range (450-880, 360-730), and after cropping and scaling, it is unified to an 800×600 pixel specification. A grayscale threshold of 85 (0-255) is set, and a grayscale feature analysis algorithm is used to scan the ROI to distinguish low-brightness areas with grayscale values ​​of 22-80. Combined with the occlusion areas detected by the photoelectric switch (X620-800mm, Y560-700mm), continuous regions with a connected region area > 60 pixels are selected through region connectivity analysis. 15 isolated noise points (single area ≤ 5 pixels) are removed, and then the Canny edge detection algorithm (threshold 55 / 155) is used to locate the closed shadow contour.

[0069] Based on the camera calibration parameters (installation height 1.4 meters, resolution 1920×1080), the actual spatial area corresponding to a single pixel is determined to be 0.045 mm². After regional pixel statistics, the total number of pixels within the shadow outline is 13200, with a preliminary calculated area of ​​13200 × 0.045 = 594 mm². After correction using an error correction algorithm (correction coefficient 0.97), the final shadow area is 576 mm².

[0070] Step S300: Based on the current actual brightness value and shadow area of ​​each LED, identify the set of LEDs that are blocked and the set of LEDs that are not blocked; for the set of LEDs that are blocked, calculate the brightness deviation between each LED and the height and angle deviation of each LED group;

[0071] Specifically, the preset reference brightness value of each LED is extracted, and a brightness deviation threshold is set; the current actual brightness value of all LEDs is traversed, and the deviation value between the actual brightness of each LED and the reference brightness is calculated. When the deviation value exceeds the preset threshold, the LED is included in the low-brightness LED candidate set; based on the shadow area, the complete spatial coordinate range of the shadow area is extracted; each LED in the low-brightness LED candidate set is traversed, and its corresponding illumination area coordinates are called; through the spatial range intersection judgment logic, it is determined whether the illumination range of the LED overlaps with the spatial range of the shadow area; LEDs whose illumination range overlaps with the shadow area are retained, LEDs without overlap are removed, forming a set of blocked LEDs, and the remaining LEDs are included in the set of unblocked LEDs.

[0072] Furthermore, for the blocked LEDs within the same blocked LED group, the average current actual brightness of all blocked LEDs in that group is used as the group's reference brightness. When cross-group comparison is required, an additional overall average brightness of the blocked LEDs is set as the global reference brightness. Each blocked LED in the same blocked LED group is iterated through, and the difference between its current actual brightness and the group's reference brightness is calculated to obtain the group's brightness deviation for a single LED. All blocked LEDs in the group are paired, and the brightness difference between each pair is calculated to form a brightness deviation matrix between LEDs in the group. The positive and negative attributes of each brightness deviation are labeled, with positive values ​​indicating that the LED's brightness is higher than the reference or comparison LED, and negative values ​​indicating that the LED's brightness is lower than the reference or comparison LED.

[0073] For each lamp group to which the obscured LED belongs, extract its preset reference height and current actual height, calculate the difference between the two, and obtain the height deviation of the lamp group; clarify the meaning of positive and negative deviations, a positive value indicates that the current height is lower than the reference height, and a negative value indicates that the current height is higher than the reference height; extract the preset reference rotation angle and current actual rotation angle of each lamp group to which the obscured LED belongs, calculate the difference between the two, and obtain the angle deviation of the lamp group; clarify the meaning of positive and negative deviations, a positive value indicates that the current rotation angle deviates from the reference angle counterclockwise, and a negative value indicates that it deviates from the reference angle clockwise.

[0074] In one specific embodiment, a preset reference brightness value is extracted for each LED: the reference brightness of 15 covering LEDs (21-28, 46-50) is set to 5500 LUX, and the reference brightness of the remaining 65 LEDs is set to 4000 LUX. A brightness deviation threshold of 500 LUX is set. The actual brightness of 80 LEDs is traversed, and a low-brightness candidate set of 10 LEDs with deviation values ​​exceeding the threshold (23-26, 47-50, 28, 46) is selected. Their brightness range is 4500-5000 LUX, and their deviation value is -1000 to -500 LUX. Combined with the spatial range of the S200 shadow area (X620-800mm, Y560-700mm, Z700-760mm), it is determined that the illumination range of the 10 candidate LEDs overlaps with the shadow area, forming a set of blocked LEDs. The remaining 70 LEDs are included in the set of unblocked LEDs.

[0075] The obscured LEDs belong to three groups: Group A (LEDs 23-26), Group B (LEDs 47-50), and Group C (LEDs 28 and 46). Group A has an average brightness of 4750 LUX (group benchmark), with individual LED deviations of -250 LUX, -50 LUX, +80 LUX, and +220 LUX, forming a deviation matrix when paired. For example, LEDs 23 and 24 have a deviation of -200 LUX, and LEDs 25 and 26 have a deviation of -140 LUX. Group B has an average brightness of 4800 LUX, with a deviation range of -300 to +250 LUX. Group C has an average brightness of 4650 LUX, with a deviation of +100 LUX between LEDs. Positive values ​​are above the benchmark, and negative values ​​are below the benchmark.

[0076] The preset reference heights A, B, and C are all 1400mm. Displacement sensors detect the current actual heights as follows: A group 1392mm, B group 1395mm, and C group 1393mm, with height deviations of +8mm, +5mm, and +7mm respectively (all currently below the reference). The rotation angles of all three references are 0°. Angle sensors detect current deviations of +1.8° for A group, -1.2° for B group, and +0.9° for C group, corresponding to counter-clockwise, clockwise, and counter-clockwise deviations respectively.

[0077] Step S400: Determine the brightness reduction ratio of the blocked LEDs based on the brightness deviation between each LED, calculate the brightness reduction value and total reduction amount of the blocked LEDs, and determine the brightness compensation parameters for the unblocked LEDs; generate a correlation function between the shadow area and the LED group attitude parameters based on the height and angle deviations of each LED group to obtain the attitude adjustment amount; generate a brightness control signal based on the brightness reduction value of the blocked LEDs and the brightness compensation parameters for the unblocked LEDs, and generate an attitude control signal based on the attitude adjustment amount to control the surgical light.

[0078] Specifically, a deviation grading algorithm is used to divide the brightness deviation between the blocked LED and other LEDs in the same group into several consecutive levels, with each level matching a preset basic brightness reduction ratio range; the current actual brightness value of each blocked LED is used as the base, multiplied by its corresponding final brightness reduction ratio, to obtain the basic brightness reduction value required for each LED; the brightness reduction values ​​of all blocked LEDs are summed to obtain the total brightness reduction of the set of blocked LEDs.

[0079] The difference between the overall target illuminance of the surgical light and the current overall actual illuminance is calculated. An overlay operation logic is used to add the basic illuminance gap to the total brightness reduction of the blocked LEDs, yielding the total compensation brightness requirement for the unblocked LEDs. A balanced allocation algorithm combined with capability adaptation is employed, considering the current actual brightness and maximum brightness carrying capacity of the unblocked LEDs, to distribute the total compensation brightness requirement to each unblocked LED. Finally, the current actual brightness value of the unblocked LEDs is multiplied by their corresponding compensation ratio to obtain the basic increase in brightness required for each LED.

[0080] Furthermore, the types of topological nodes are determined. Input nodes include shadow area, height deviation, and angle deviation; output nodes include target height and rotation angle of the light group; intermediate nodes include target coordinates and illumination range of the light group. Edges represent the correlation strength between nodes, and a basic framework of the topological structure is built. The correlation degree between each node is calculated through a correlation analysis algorithm, and nodes with a correlation degree higher than a preset threshold are connected with edges to form a topological correlation network that includes the influence of posture deviation. A topology optimization algorithm is used to remove redundant edges with too low correlation strength and merge duplicate nodes.

[0081] A topology analysis algorithm is used to identify the dependency paths of each node in the topology network and determine the transmission logic. Specifically, using the shadow area as the core response variable and considering the effects of height and angle deviations, the key transmission path from the input node to the output node is derived, and the association rules and quantification parameters on the path are extracted. Based on the transmission paths and association strengths of the topology nodes, and combined with a data fitting algorithm, the relationship between the shadow area and attitude deviation and the target attitude parameters of the lamp group is transformed into a function, forming a preliminary association function. Historical valid data samples are substituted into the preliminary function, and an error analysis algorithm is used to calculate the deviation between the target attitude parameters output by the function and the actual data. When the deviation exceeds the allowable threshold, the process returns to the topology network optimization step to adjust the parameters and regenerate the function until the deviation meets the requirements, thus determining the association function between the shadow area and the attitude parameters of the lamp group.

[0082] Substitute the real-time shadow area, real-time height deviation, and real-time angle deviation data of the current surgery into the function, and obtain the target posture parameters of the lamp group required to reduce the shadow area to the preset allowable range through function calculation; calculate the difference between the current posture parameters of the lamp group and the target posture parameters output by the associated function to obtain the height adjustment amount and the rotation angle adjustment amount, and mark the direction of the adjustment amount.

[0083] In one specific embodiment, a deviation grading algorithm is used to divide the data into three levels, corresponding to the basic reduction ratio range: Level 1 (5%-7%) with an absolute deviation value ≤1000LUX, Level 2 (7%-10%) with a deviation value of 1000-2500LUX, and Level 3 (10%-12%) with a deviation value >2500LUX. Based on the brightness deviations of each group, the final allocation ratios are as follows: Group A: No. 23 (-2500 LUX, Level 2) 9%, No. 24 (-500 LUX, Level 1) 6%, No. 25 (+800 LUX, Level 1) 5%, No. 26 (+2200 LUX, Level 2) 8%; Group B: No. 47 (-3000 LUX, Level 3) 11%, No. 48 (-1200 LUX, Level 2) 7%, No. 49 (+1800 LUX, Level 2) 9%, No. 50 (+2500 LUX, Level 2) 10%; Group C: No. 28 (+400 LUX, Level 1) 5%, No. 46 (-600 LUX, Level 1) 6%. Based on the current brightness, the total brightness reduction is 14200 LUX for Group A, 15800 LUX for Group B, and 1680 LUX for Group C, for a total brightness reduction of 31680 LUX.

[0084] The preset overall target illuminance is 150,000 LUX, the current actual overall illuminance is 132,000 LUX, and the basic illuminance gap is 18,000 LUX. After adding the total reduction, the total compensation requirement for the unblocked LEDs is 49,680 LUX. The maximum brightness carrying capacity of the 70 unblocked LEDs is 160,000 LUX each. Based on the current brightness (800-95,000 LUX), the compensation ratio is allocated (3%-6%), with 5%-6% allocated to low-brightness LEDs and 3%-4% to the rest. The calculated brightness increase per LED ranges from 24 to 5,700 LUX, and the brightness after compensation does not exceed the standard.

[0085] When constructing the topology, a correlation strength threshold of 0.72 was set. Correlation analysis showed that the correlation strength between the shadow area and the height deviation of group A was 0.83, and the correlation strength with the angle deviation of group B was 0.79. The correlation strength between the target coordinates and the illumination range was 0.90. After removing redundant edges, an optimized network was formed. A preliminary correlation function was generated by fitting data, and verified using 60 sets of historical data. The initial error was 10%, which was reduced to 3.5% after optimization (below the allowable threshold of 5%), thus determining the final correlation function.

[0086] Substituting the real-time shadow area of ​​576mm² and the attitude deviations of each group, the target attitude parameters are calculated as follows: Group A: target height 1408mm, rotation angle 0°; Group B: target height 1405mm, rotation angle 0°; Group C: target height 1407mm, rotation angle 0°. Adjustments are calculated as follows: Group A: height +16mm (increase), angle -1.8° (clockwise correction); Group B: height +10mm (increase), angle +1.2° (counterclockwise correction); Group C: height +14mm (increase), angle -0.9° (clockwise correction). Corresponding brightness control signals are generated (current decreases of 0.015-0.022A for blocked LEDs, current increases of 0.002-0.005A for unblocked LEDs) and attitude control signals, driving the adjustment execution.

[0087] like Figure 2 As shown, this application provides an environmental sensing-based surgical lighting control system, including:

[0088] Target localization and LED bead set matching module: including: target coordinate extraction unit acquires images of the surgical site and obtains target coordinates and boundary coordinate point set through feature matching; LED bead set matching unit determines the LED bead set covering the coordinate region jointly defined by the target coordinates and boundary coordinate point set based on the target coordinates and matching mapping relationship library.

[0089] The shadow area calculation module includes: a lamp brightness inverse calculation unit that collects the occlusion signal and brightness signal of each area in the surgical lighting area and inversely calculates the current actual brightness value of each lamp; and a shadow area calculation unit that extracts the shadow area formed on the target coordinates after the doctor blocks the light based on the target coordinates and boundary coordinate point set, and calculates the shadow area.

[0090] Deviation calculation module: includes: a shading and unshading set identification unit that identifies the shading set and the unshading set of lamp beads based on the current actual brightness value and shadow area of ​​each lamp bead; and a deviation calculation unit that calculates the brightness deviation between each lamp bead and the height and angle deviation of each lamp group for the shading set of lamp beads.

[0091] The control execution module includes: a brightness adjustment parameter calculation unit that determines the brightness reduction ratio of the blocked LEDs based on the brightness deviation between each LED, calculates the required brightness reduction value and total reduction amount for the blocked LEDs, and determines the brightness compensation parameters for the unblocked LEDs; a posture adjustment parameter calculation unit that generates a correlation function between the shadow area and the posture parameters of each light group based on the height and angle deviations of each light group, and obtains the posture adjustment amount; and a control execution unit that generates a brightness control signal based on the required brightness reduction value of the blocked LEDs and the brightness compensation parameters for the unblocked LEDs, and generates a posture control signal based on the posture adjustment amount to control the surgical lights.

[0092] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A surgical lighting control method based on environmental sensing, characterized in that, Includes the following steps: Images of the surgical site are acquired, and the target coordinates and boundary coordinate point set are obtained through feature matching. Based on the target coordinates, a matching mapping database is used to determine the set of LED beads covering the coordinate region jointly defined by the target coordinates and boundary coordinate point set. Collect the occlusion signal and brightness signal of the surgical lighting area, and calculate the current actual brightness value of each lamp; Based on the target coordinates and boundary coordinates, the shadow area formed on the target coordinates after the doctor blocks the light is extracted, and the shadow area is calculated. The process involves collecting occlusion signals and brightness signals from the surgical lighting area, and then calculating the current actual brightness value of each LED, including: Brightness data is collected from the lighting areas according to preset zones to obtain the brightness signals of each zone; A convolutional neural network is used to train a model of the mapping relationship between lamp current and brightness. Combined with the digital brightness signal, the current power supply current signal of the corresponding area lamp group is inferred. According to the relationship between the lamp group and the lamp, the current power supply current signal is allocated to the individual lamp, and the actual brightness value of each lamp is calculated by substituting it into the mapping relationship model. Based on the current actual brightness value and shadow area of ​​each LED, identify the set of LEDs that are blocked and the set of LEDs that are not blocked; for the set of LEDs that are blocked, calculate the brightness deviation between each LED and the height and angle deviation of each LED group; Based on the brightness deviation between each LED, the brightness reduction ratio of the blocked LED is determined, the required brightness reduction value and total reduction amount of the blocked LED are calculated, and the brightness compensation parameters of the unblocked LED are determined. Based on the height and angle deviations of each LED group, a correlation function between the shadow area and the LED group attitude parameters is generated to obtain the attitude adjustment amount. Based on the required brightness reduction value of the blocked LED and the brightness compensation parameters of the unblocked LED, a brightness control signal is generated, and based on the attitude adjustment amount, an attitude control signal is generated to control the surgical light.

2. The surgical lighting control method based on environmental sensing according to claim 1, characterized in that, The acquired images of the surgical site are used to obtain a set of target coordinates and boundary coordinate points through feature matching, including: The image acquisition camera array is started, and the image recognition parameters are initialized. The camera array continuously acquires images of the surgical operation area according to the initialized frame rate, obtaining a raw image sequence containing the surgical site and the surrounding environment. Image preprocessing is performed on the acquired raw image sequence, including grayscale conversion, Gaussian filtering for noise reduction, and image enhancement, and the preprocessed image set is output. Based on the preset lesion feature template, a feature matching algorithm is executed on the preprocessed image set. By calculating the similarity between each region in the image and the feature template, regions with similarity greater than the feature matching threshold are selected as lesion regions, and the pixel coordinate point set of the lesion region is obtained. This set is then converted into a three-dimensional spatial dimension to obtain spatial coordinate data. Combined with the position markers of each camera, the spatial coordinate data collected by multiple cameras are fused to output the target coordinates and boundary coordinate point set.

3. The surgical lighting control method based on environmental sensing according to claim 1, characterized in that, The process of determining the set of LED beads covering the coordinate region defined by the target coordinates and the boundary coordinate point set, based on the target coordinates and a matching mapping database, includes: The system calls a mapping library containing 3D coordinate partitioning rules for the surgical lighting area, illumination characteristic parameters of individual LEDs, and preset associations between coordinate sub-partitions and LEDs. The target coordinates and boundary coordinate point sets are substituted one by one into the 3D coordinate partitioning rules. When the X, Y, and Z axis values ​​of a coordinate point all fall within the coordinate threshold range of a certain sub-partition, the coordinate point is determined to belong to that sub-partition. All sub-partitions to which coordinate points belong are summarized to form a partition set. The preset associations between coordinate sub-partitions and LEDs are called to extract all LEDs corresponding to each sub-partition in the partition set, forming a candidate LED list L. Obtain the actual irradiation distance between the surgical light and the affected area, and call the irradiation characteristic parameters of each lamp bead in L; for each lamp bead in L, calculate its effective irradiation range at the actual irradiation distance based on the relationship between the lamp bead's irradiation angle, irradiation distance and coverage area, and determine whether the effective irradiation range intersects with the target coordinate area; retain the lamp beads that are determined to have intersection, and remove the lamp beads that do not intersect, forming a set of lamp beads that covers the coordinate area jointly defined by the target coordinates and the boundary coordinate point set.

4. The surgical lighting control method based on environmental sensing according to claim 1, characterized in that, The process involves collecting occlusion signals and brightness signals from the surgical lighting area, and then calculating the current actual brightness value of each LED, including: The photoelectric switch sensor array scans the surgical illumination area at a preset frequency to capture the occlusion signal generated by the obstruction. The occlusion signal and the brightness signal are denoised and filtered to remove interference signals below the preset intensity threshold. The signals are then input into an A / D converter, which converts the analog signal into a digital signal according to the initial sampling accuracy and conversion rate, thus obtaining the occlusion digital signal and the brightness digital signal.

5. The surgical lighting control method based on environmental sensing according to claim 1, characterized in that, The process of extracting the shadow region formed on the target coordinates after the doctor blocks the light, based on the target coordinates and boundary coordinate point set, and calculating the shadow area includes: Based on the target coordinates and boundary coordinate point set, the spatial region corresponding to the target coordinates is defined in the image to generate the target region ROI. The ROI represents the region of interest, and the background region in the image that is not related to the surgical site is excluded. The target region ROI is then cropped and scaled to unify the image size specifications. A grayscale feature analysis algorithm is used to perform a full-domain scan of the pixel grayscale values ​​within the target region ROI to distinguish low-brightness regions with grayscale values ​​below a preset threshold. Combining the spatial correspondence between the shadow region and the location of the occluder, a region connectivity analysis algorithm is used to filter out low-brightness regions with continuous boundaries and whose locations match the occluder regions detected by the photoelectric switch sensor, thus eliminating isolated low-brightness noise points. An edge detection algorithm is used to locate the complete boundary of the filtered shadow region, forming a closed shadow region outline. Based on the calibration parameters of the image acquisition camera, the actual spatial area corresponding to a single pixel in the image is determined. A regional pixel statistics algorithm is used to count all pixels within the extracted shadow region contour to obtain the total number of pixels corresponding to the shadow region. Combining the actual spatial area corresponding to a single pixel, the total number of pixels in the shadow region is converted into the actual spatial area through multiplication. At the same time, an error correction algorithm is used to correct the conversion result to obtain the shadow area.

6. The surgical lighting control method based on environmental sensing according to claim 1, characterized in that, The process of identifying the set of blocked LEDs and the set of unblocked LEDs based on the current actual brightness value and shadow area of ​​each LED includes: Extract the preset reference brightness value of each LED and set a brightness deviation threshold; iterate through the current actual brightness value of all LEDs and calculate the deviation value between the actual brightness of each LED and the reference brightness. When the deviation value exceeds the preset threshold, the LED is included in the low brightness LED candidate set; based on the shadow area, extract the complete spatial coordinate range of the shadow area; iterate through each LED in the low brightness LED candidate set and call its corresponding illumination area coordinates; through the spatial range intersection judgment logic, determine whether the illumination range of the LED overlaps with the spatial range of the shadow area; retain LEDs whose illumination range overlaps with the shadow area, remove LEDs that do not overlap, form the set of blocked LEDs, and include the remaining LEDs in the set of unblocked LEDs.

7. The surgical lighting control method based on environmental sensing according to claim 1, characterized in that, The calculation of brightness deviations between individual LEDs and height and angle deviations between LED groups for the set of blocked LEDs includes: For each blocked LED in the same blocked LED group, the average current actual brightness of all blocked LEDs in the group is used as the group's reference brightness. When cross-group comparison is required, an additional average brightness of the blocked LEDs is set as the global reference brightness. Each blocked LED in the same blocked LED group is iterated through, and the difference between its current actual brightness and the group's reference brightness is calculated to obtain the group's brightness deviation for a single LED. All blocked LEDs in the group are paired, and the brightness difference between each pair is calculated to form a brightness deviation matrix between LEDs in the group. The positive and negative attributes of each brightness deviation are labeled, with positive values ​​indicating that the LED's brightness is higher than the reference or comparison LED, and negative values ​​indicating that the LED's brightness is lower than the reference or comparison LED. For each lamp group to which the obscured LED belongs, extract its preset reference height and current actual height, calculate the difference between the two, and obtain the height deviation of the lamp group; clarify the meaning of positive and negative deviations, a positive value indicates that the current height is lower than the reference height, and a negative value indicates that the current height is higher than the reference height; extract the preset reference rotation angle and current actual rotation angle of each lamp group to which the obscured LED belongs, calculate the difference between the two, and obtain the angle deviation of the lamp group; clarify the meaning of positive and negative deviations, a positive value indicates that the current rotation angle deviates from the reference angle counterclockwise, and a negative value indicates that it deviates from the reference angle clockwise.

8. The surgical lighting control method based on environmental sensing according to claim 1, characterized in that, The process of determining the brightness reduction ratio of the blocked LEDs based on the brightness deviation between each LED, calculating the required brightness reduction value and total reduction amount for the blocked LEDs, and determining the brightness compensation parameters for the unblocked LEDs includes: A deviation grading algorithm is used to divide the brightness deviation between the blocked LED and other LEDs in the same group into several consecutive levels. Each level is matched with a preset basic brightness reduction ratio range. The current actual brightness value of each blocked LED is used as the base, and multiplied by its corresponding final brightness reduction ratio to obtain the basic brightness reduction value of each LED. The brightness reduction values ​​of all blocked LEDs are summed to obtain the total brightness reduction of the set of blocked LEDs. The difference between the overall target illuminance of the surgical light and the current overall actual illuminance is calculated. An overlay operation logic is used to add the basic illuminance gap to the total brightness reduction of the blocked LEDs, yielding the total compensation brightness requirement for the unblocked LEDs. A balanced allocation algorithm combined with capability adaptation is employed, considering the current actual brightness and maximum brightness carrying capacity of the unblocked LEDs, to distribute the total compensation brightness requirement to each unblocked LED. Finally, the current actual brightness value of the unblocked LEDs is multiplied by their corresponding compensation ratio to obtain the basic increase in brightness required for each LED.

9. The surgical lighting control method based on environmental sensing according to claim 1, characterized in that, The correlation function between the shadow area and the lamp group attitude parameters, generated based on the height and angle deviations of each lamp group, is used to obtain the attitude adjustment amount, including: The topology node types are determined. Input nodes include shadow area, height deviation, and angle deviation. Output nodes include target height and rotation angle of the light group. Intermediate nodes include target coordinates and illumination range of the light group. Edges represent the correlation strength between nodes, and the basic framework of the topology structure is built. The correlation degree between each node is calculated through a correlation analysis algorithm. Nodes with a correlation degree higher than a preset threshold are connected with edges to form a topology network that includes the influence of posture deviation. A topology optimization algorithm is used to remove redundant edges with too low correlation strength and merge duplicate nodes. A topology analysis algorithm is used to identify the dependency paths of each node in the topology network and determine the transmission logic. Specifically, using the shadow area as the core response variable and considering the effects of height and angle deviations, the key transmission path from the input node to the output node is derived, and the association rules and quantification parameters on the path are extracted. Based on the transmission paths and association strengths of the topology nodes, and combined with a data fitting algorithm, the relationship between the shadow area and attitude deviation and the target attitude parameters of the lamp group is transformed into a function, forming a preliminary association function. Historical valid data samples are substituted into the preliminary function, and an error analysis algorithm is used to calculate the deviation between the target attitude parameters output by the function and the actual data. When the deviation exceeds the allowable threshold, the process returns to the topology network optimization step to adjust the parameters and regenerate the function until the deviation meets the requirements, thus determining the association function between the shadow area and the attitude parameters of the lamp group. Substitute the real-time shadow area, real-time height deviation, and real-time angle deviation data of the current surgery into the function, and obtain the target posture parameters of the lamp group required to reduce the shadow area to the preset allowable range through function calculation; calculate the difference between the current posture parameters of the lamp group and the target posture parameters output by the associated function to obtain the height adjustment amount and the rotation angle adjustment amount, and mark the direction of the adjustment amount.

10. An environmental sensing-based surgical lighting control system, using the environmental sensing-based surgical lighting control method according to any one of claims 1-9, characterized in that, include: Target localization and LED bead set matching module: including: target coordinate extraction unit acquires images of the surgical site and obtains target coordinates and boundary coordinate point set through feature matching; LED bead set matching unit determines the LED bead set covering the coordinate region jointly defined by the target coordinates and boundary coordinate point set based on the target coordinates and matching mapping relationship library. The shadow area calculation module includes: a lamp brightness inverse calculation unit that collects the occlusion signal and brightness signal of each area in the surgical lighting area and inversely calculates the current actual brightness value of each lamp; and a shadow area calculation unit that extracts the shadow area formed on the target coordinates after the doctor blocks the light based on the target coordinates and boundary coordinate point set, and calculates the shadow area. Deviation calculation module: includes: a shading and unshading set identification unit that identifies the shading set and the unshading set of lamp beads based on the current actual brightness value and shadow area of ​​each lamp bead; and a deviation calculation unit that calculates the brightness deviation between each lamp bead and the height and angle deviation of each lamp group for the shading set of lamp beads. The control execution module includes: a brightness adjustment parameter calculation unit that determines the brightness reduction ratio of the blocked LEDs based on the brightness deviation between each LED, calculates the required brightness reduction value and total reduction amount for the blocked LEDs, and determines the brightness compensation parameters for the unblocked LEDs; a posture adjustment parameter calculation unit that generates a correlation function between the shadow area and the posture parameters of each light group based on the height and angle deviations of each light group, and obtains the posture adjustment amount; and a control execution unit that generates a brightness control signal based on the required brightness reduction value of the blocked LEDs and the brightness compensation parameters for the unblocked LEDs, and generates a posture control signal based on the posture adjustment amount to control the surgical lights.

Images