Machine vision light source illumination intensity adaptive correction method and system

Abstract

This invention discloses a method and system for adaptive correction of illumination intensity of a machine vision light source. The method extracts pixel frequency domain distribution features from a reference image of a target object under reference illumination conditions to calibrate anomalous sub-regions of surface illumination on the target object. Based on the illumination energy distribution of these anomalous sub-regions, an appropriate illumination state is determined. By comparing the reference illumination conditions and the appropriate illumination state, the actual illumination parameters of the anomalous sub-regions are determined. Based on the layout of all anomalous sub-regions and their actual illumination parameters, the illumination intensity variation pattern of the target object during the movement of the machine vision light source is adjusted. By extracting the pixel frequency domain distribution under reference illumination, anomalous sub-regions where reference illumination cannot ensure the acquisition of structural detail information are calibrated. Adaptive intensity variation illumination is implemented for all anomalous sub-regions to avoid obscuring structural details and improve the accuracy of machine vision recognition.

Description

Technical Field

[0001] This invention relates to the field of machine vision, and in particular to a method and system for adaptive correction of illumination intensity of machine vision light sources. Background Technology

[0002] Machine vision integrates optics, electromechanical systems, and computer vision technologies. It utilizes corresponding processing algorithms to extract useful information from images, thereby meeting various needs such as defect detection and target tracking. In machine vision scenarios, to compensate for deficiencies such as insufficient or uneven lighting in the natural environment, appropriate illumination is applied to the target object using a machine vision light source to ensure accurate and comprehensive capture of the object's structural details during image acquisition. Considering the complex and varied surface contours of target objects, with different surface areas possessing different dimensions and contour structures, and varying light reflection states, applying uniform intensity illumination to the target object's surface using a machine vision light source may lead to overexposure of contour details in some areas, preventing them from being properly displayed in the image and affecting the accuracy of machine vision recognition. Therefore, adaptively adjusting the illumination intensity of the machine vision light source according to the contour structure of different areas of the target object's surface is of great significance for avoiding the obscuring of contour details and improving the accuracy and reliability of machine vision recognition. Summary of the Invention

[0003] Considering that existing machine vision light sources apply uniform intensity illumination to the target object, the contour details of some areas on the target object's surface cannot be fully and correctly presented in the image, resulting in missing local surface contour information and reducing the accuracy and reliability of global machine vision recognition of the target object's surface. In view of the above problems, this invention provides an adaptive correction method for the illumination intensity of machine vision light sources, including: A reference image of the target object under reference lighting conditions is obtained, and the pixel frequency domain distribution features of the reference image are extracted; based on the pixel frequency domain distribution features, the surface lighting aberration sub-regions of the target object are identified. Based on the light energy distribution of the surface illumination anomaly sub-region, determine the appropriate illumination state for the surface illumination anomaly sub-region; compare the reference illumination conditions and the appropriate illumination state to determine the actual illumination parameters of the surface illumination anomaly sub-region. Based on the layout of all surface lighting anomaly sub-regions and the actual lighting parameters, adjust the lighting intensity variation pattern of the target object during the movement of the machine vision light source.

[0004] Optionally, a reference image of the target object under reference lighting conditions is acquired, and pixel frequency domain distribution features of the reference image are extracted; based on the pixel frequency domain distribution features, the surface illumination anomalous sub-regions of the target object are identified, including: Acquire at least two reference images of a target object under reference uniform illumination along at least two directions; wherein the incident angle of the light rays of the reference uniform illumination relative to the surface of the target object satisfies a preset angle range, and the reference uniform illumination along at least two directions is applied alternately to the surface of the target object in time; The reference image is subjected to pixel frequency domain transformation to obtain the pixel contour frequency domain components of the reference image, and the average proportion of the pixel contour high-frequency components of each grid sub-region under all reference images is extracted. A threshold comparison is performed on the average proportion of high-frequency components of the pixel contour in each grid sub-region to determine whether the grid sub-region belongs to a surface illumination anomalous sub-region.

[0005] Optionally, based on the illumination energy distribution of the surface illumination anomaly sub-region, an adapted illumination state for the surface illumination anomaly sub-region is determined; by comparing the reference illumination conditions and the adapted illumination state, the actual illumination parameters of the surface illumination anomaly sub-region are determined, including: Frequency domain analysis is performed on the global pixel brightness of the surface illumination anomaly sub-region to obtain the frequency domain distribution of illumination energy; based on the frequency domain distribution of illumination energy, the overexposure degree of the surface illumination anomaly sub-region is determined, thereby obtaining the appropriate illumination intensity; By comparing the reference illumination intensity and the adapted illumination intensity obtained by the surface illumination anomaly sub-region under the reference illumination conditions, an illumination intensity difference is obtained; based on the illumination intensity difference, the actual illumination intensity parameters of the surface illumination anomaly sub-region are determined.

[0006] Optionally, based on the layout of all surface illumination anomaly sub-regions and the actual illumination parameters, the illumination intensity variation pattern of the target object during the movement of the machine vision light source is adjusted, including: Based on the layout of all surface illumination anomaly sub-regions on the target object surface, determine the timing of illumination intensity changes during machine vision light source movement; Based on the actual illumination parameters, determine the target value of illumination intensity change during the movement of the machine vision light source; Based on the timing of the light intensity change and the target value of the light intensity change, the light intensity change pattern of the target object during the movement of the machine vision light source is adjusted.

[0007] As one aspect of the present invention, embodiments of the present invention also provide an adaptive correction system for the illumination intensity of a machine vision light source, comprising: The reference image processing module is used to acquire a reference image of the target object under reference lighting conditions and extract the pixel frequency domain distribution features of the reference image; An illumination anomaly calibration module is used to calibrate the surface illumination anomaly sub-region of the target object based on the pixel frequency domain distribution characteristics; An adaptive illumination determination module is used to determine the adaptive illumination state of the abnormal surface illumination sub-region based on the illumination energy distribution of the abnormal surface illumination sub-region. The illumination parameter determination module is used to compare the reference illumination conditions and the adapted illumination state to determine the actual illumination parameters of the surface illumination anomaly sub-region. The illumination mode adjustment module is used to adjust the illumination intensity change mode of the target object during the movement of the machine vision light source based on the layout of all surface illumination abnormal sub-regions and the actual illumination parameters.

[0008] Optionally, the reference image processing module is used to acquire a reference image of the target object under reference illumination conditions, and extract the pixel frequency domain distribution features of the reference image, including: Acquire at least two reference images of a target object under reference uniform illumination along at least two directions; wherein the incident angle of the light rays of the reference uniform illumination relative to the surface of the target object satisfies a preset angle range, and the reference uniform illumination along at least two directions is applied alternately to the surface of the target object in time; The reference image is subjected to pixel frequency domain transformation to obtain the pixel contour frequency domain components of the reference image, and the average proportion of the pixel contour high-frequency components of each grid sub-region under all reference images is extracted. The illumination anomaly calibration module is used to calibrate the surface illumination anomaly sub-regions of the target object based on the pixel frequency domain distribution characteristics, including: A threshold comparison is performed on the average proportion of high-frequency components of the pixel contour in each grid sub-region to determine whether the grid sub-region belongs to a surface illumination anomalous sub-region.

[0009] Optionally, the adaptive illumination determination module is used to determine the adaptive illumination state of the surface illumination anomaly sub-region based on the illumination energy distribution of the surface illumination anomaly sub-region, including: Frequency domain analysis is performed on the global pixel brightness of the surface illumination anomaly sub-region to obtain the frequency domain distribution of illumination energy; based on the frequency domain distribution of illumination energy, the overexposure degree of the surface illumination anomaly sub-region is determined, thereby obtaining the appropriate illumination intensity; The illumination parameter determination module is used to compare the baseline illumination conditions and the adapted illumination state to determine the actual illumination parameters of the surface illumination anomaly sub-region, including: By comparing the reference illumination intensity and the adapted illumination intensity obtained by the surface illumination anomaly sub-region under the reference illumination conditions, an illumination intensity difference is obtained; based on the illumination intensity difference, the actual illumination intensity parameters of the surface illumination anomaly sub-region are determined.

[0010] Optionally, the illumination mode adjustment module is used to adjust the illumination intensity change mode of the target object during the movement of the machine vision light source based on the layout of all surface illumination anomaly sub-regions and the actual illumination parameters, including: Based on the layout of all surface illumination anomaly sub-regions on the target object surface, determine the timing of illumination intensity changes during machine vision light source movement; Based on the actual illumination parameters, determine the target value of illumination intensity change during the movement of the machine vision light source; Based on the timing of the light intensity change and the target value of the light intensity change, the light intensity change pattern of the target object during the movement of the machine vision light source is adjusted.

[0011] The beneficial effects of the above-mentioned technical solutions provided in the embodiments of the present invention include at least the following: This invention provides a method and system for adaptive correction of illumination intensity of a machine vision light source. The method extracts pixel frequency domain distribution features from a reference image of a target object under reference illumination conditions to calibrate abnormal surface illumination sub-regions of the target object. Based on the illumination energy distribution of these sub-regions, an appropriate illumination state is determined. By comparing the reference illumination conditions and the appropriate illumination state, the actual illumination parameters of the abnormal surface illumination sub-regions are determined. Based on the layout of all abnormal surface illumination sub-regions and their actual illumination parameters, the illumination intensity variation pattern of the target object during the movement of the machine vision light source is adjusted. By extracting the pixel frequency domain distribution under reference illumination, abnormal sub-regions where the reference illumination cannot ensure the acquisition of structural detail information are calibrated. Adaptive intensity variation illumination is implemented for all abnormal sub-regions to avoid obscuring structural details and improve machine vision recognition accuracy.

[0012] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0013] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0014] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the adaptive correction method for illumination intensity of a machine vision light source provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the machine vision light source illumination intensity adaptive correction system provided in an embodiment of the present invention. Detailed Implementation

[0015] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0016] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," "far," "near," "front," and "rear," etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings and are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0017] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0018] Please see Figure 1 As shown, an embodiment of this application provides a method for adaptive correction of illumination intensity of a machine vision light source. This method includes: Acquire a reference image of the target object under reference lighting conditions, and extract the pixel frequency domain distribution features of the reference image; based on the pixel frequency domain distribution features, identify the surface lighting aberration sub-regions of the target object; Based on the illumination energy distribution of the surface illumination anomaly sub-region, determine the appropriate illumination state for the surface illumination anomaly sub-region; compare the reference illumination conditions and the appropriate illumination state to determine the actual illumination parameters of the surface illumination anomaly sub-region. Based on the layout and actual lighting parameters of all surface lighting anomaly sub-regions, adjust the lighting intensity variation pattern of the target object during the movement of the machine vision light source.

[0019] The beneficial effects of the above embodiments are that the machine vision light source illumination intensity adaptive correction method extracts the pixel frequency domain distribution under reference illumination, calibrates abnormal sub-regions where the reference illumination cannot ensure the acquisition of structural detail information, and implements adaptive intensity variation illumination for all abnormal sub-regions to avoid obscuring structural details and improve machine vision recognition accuracy.

[0020] In another embodiment, a reference image of the target object under reference lighting conditions is acquired, and pixel frequency domain distribution features of the reference image are extracted; based on the pixel frequency domain distribution features, the surface illumination anomalous sub-regions of the target object are identified, including: Acquire at least two reference images of the target object under reference uniform illumination in at least two directions; wherein the incident angle of the light rays of the reference uniform illumination relative to the surface of the target object satisfies a preset angle range, and the reference uniform illumination in at least two directions is applied alternately to the surface of the target object in time; The pixel frequency domain transformation of the reference image is performed to obtain the pixel contour frequency domain components of the reference image, and the average proportion of the pixel contour high-frequency components of each grid sub-region under all reference images is extracted. A threshold comparison is performed on the average proportion of high-frequency components of the pixel contour in each grid sub-region to determine whether the grid sub-region belongs to the abnormal surface illumination sub-region.

[0021] The beneficial effects of the above embodiments are that when a machine vision light source illuminates the surface of a target object, the surface reflects the light. The overall contour microstructure of the target object's surface is not uniform; some surface areas exhibit gentle contour microstructure changes, while others show complex changes. This results in different surface areas having different light reflection (e.g., diffuse reflection) characteristics. These different light reflection characteristics directly affect the degree to which the contour details of the corresponding surface areas are captured during image acquisition. It is understandable that when a surface area strongly reflects light, it will be overexposed during image acquisition, obscuring its contour details. Conversely, when a surface area reflects light moderately, it will be captured normally, effectively presenting its contour details in the image. The high-frequency components of pixel contours in the image correspond to the contour details of the target object's surface. The more high-frequency components of pixel contours in the image, the more contour details are captured during image acquisition. Based on the relationship between the illumination of the target object by the machine vision light source, the proportion of high-frequency components of pixel contours in the image, and the contour details of the target object's surface, the actual illumination state of the target object's surface can be accurately determined by identifying the high-frequency components of pixel contours in the image.

[0022] In practice, to accurately determine the high-frequency components of the pixel contour after imaging the target object, reference light sources are arranged on at least two different sides of the target object's surface. These reference light sources can be, but are not limited to, LED light sources, and each reference light source can independently illuminate the target object's surface. All reference light sources can be controlled to alternately illuminate the target object's surface with uniform intensity over time, ensuring that the entire target object's surface receives uniform reference illumination. Furthermore, the light emitted by each reference light source can be incident on the target object's surface in a grazing incidence manner; for example, the angle of incidence of the light emitted by each reference light source relative to the target object's surface can be greater than 160°. During the illumination of the target object by each reference light source, a reference image of the target object's surface is acquired, and each reference image comprehensively records the structural information of the target object's surface. Pixel frequency domain transformation is performed on each reference image to obtain the pixel contour frequency domain components of the global range of the reference image. The target object's surface is then divided into several grid sub-regions. Using these grid sub-regions as units, the corresponding high-frequency components of the pixel contour are extracted from the pixel contour frequency domain components, and the proportion of the high-frequency components of the pixel contour in each grid sub-region is calculated (i.e., the proportion of the high-frequency components of the pixel contour in the pixel contour frequency domain components within each grid sub-region). Furthermore, for each grid sub-region within the global scope of the target object's surface, the proportion of the high-frequency components of the pixel contour in all reference images for each grid sub-region is obtained, and the average proportion of the high-frequency components of the pixel contour in each grid sub-region is calculated. The average proportion of the high-frequency components of the pixel contour in each grid sub-region is compared with a preset proportion threshold. If the average proportion of the high-frequency components of the pixel contour in each grid sub-region is less than the preset proportion threshold, then the grid sub-region is labeled as a surface illumination anomaly sub-region; otherwise, it is not labeled as a surface illumination anomaly sub-region. It can be understood that the aforementioned surface illumination anomaly sub-region refers to a grid sub-region where the structural details of the target object's surface cannot be properly displayed in the image due to abnormal illumination (such as defects such as excessively high or low light intensity).

[0023] In another embodiment, based on the illumination energy distribution of the surface illumination anomaly sub-region, the adapted illumination state of the surface illumination anomaly sub-region is determined; by comparing the baseline illumination conditions and the adapted illumination state, the actual illumination parameters of the surface illumination anomaly sub-region are determined, including: Frequency domain analysis is performed on the global pixel brightness of the surface illumination anomalous sub-region to obtain the frequency domain distribution of illumination energy; based on the frequency domain distribution of illumination energy, the overexposure degree of the surface illumination anomalous sub-region is determined, thereby obtaining the appropriate illumination intensity; By comparing the baseline illumination intensity and the adapted illumination intensity obtained under the baseline illumination conditions in the surface illumination anomaly sub-region, the illumination intensity difference is obtained; based on the illumination intensity difference, the actual illumination intensity parameters of the surface illumination anomaly sub-region are determined.

[0024] Understandably, in sub-regions with abnormal surface lighting, the higher the pixel brightness, the greater the degree to which the corresponding structural details are overexposed and obscured, making it impossible to accurately identify the true structural details of the target object at these locations. In such cases, adjusting the illumination intensity at these locations is necessary to suppress overexposure of structural details. To accurately calibrate the degree of overexposure in sub-regions with abnormal surface lighting, a frequency domain analysis is performed on the global pixel brightness of these sub-regions to obtain the frequency domain distribution of illumination energy, thereby determining the degree of overexposure. For example, the degree of overexposure can be obtained by inverting the frequency domain distribution of illumination energy. Then, based on the comparison between the degree of overexposure and the preset normal exposure level, an appropriate illumination intensity is obtained. This appropriate illumination intensity refers to the illumination intensity that can prevent overexposure in sub-regions with abnormal surface lighting and ensure that all structural information in these sub-regions is captured correctly.

[0025] Furthermore, by comparing the baseline illumination intensity and the adapted illumination intensity obtained under the baseline illumination conditions in the surface illumination anomaly sub-region, the illumination intensity difference is obtained. Based on the above illumination intensity difference, the actual illumination intensity parameters of the surface illumination anomaly sub-region are determined, providing an accurate basis for subsequent adjustment of the light irradiation intensity of the machine vision light source on the surface illumination anomaly sub-region.

[0026] In another embodiment, adjusting the illumination intensity variation pattern of the target object during the movement of the machine vision light source, based on the layout and actual illumination parameters of all surface illumination anomalous sub-regions, includes: Based on the layout of all surface illumination anomaly sub-regions on the target object surface, determine the timing of illumination intensity changes during machine vision light source movement; Based on the actual illumination parameters, determine the target value of illumination intensity change during the movement of the machine vision light source; Adjust the illumination intensity change pattern of the target object during the movement of the machine vision light source based on the timing of the illumination intensity change and the target value of the illumination intensity change.

[0027] It is understandable that the target object's surface has multiple sub-regions with abnormal surface lighting. To achieve effective and reliable illumination across the entire target object surface, the machine vision light source needs to apply targeted and appropriate lighting intensities to all sub-regions with abnormal surface lighting. In practice, based on the layout of all sub-regions with abnormal surface lighting on the target object's surface, the timing of lighting intensity changes during the movement of the machine vision light source is determined. Specifically, this involves the machine vision light source changing its lighting intensity on the sub-regions when it moves to an alignment position with them. Based on the actual lighting parameters, the target value for lighting intensity changes during the movement of the machine vision light source is determined. Then, based on the timing of the lighting intensity changes and the target value, the lighting intensity change pattern of the machine vision light source on the target object during its movement is adjusted. This allows the machine vision light source to change its lighting intensity state according to the alignment relationship between itself and the sub-regions with abnormal surface lighting, ensuring adaptive intensity-changing illumination for all sub-regions with abnormal surface lighting.

[0028] In another alternative embodiment, the preset movement path of the machine vision light source is regarded as a continuous spatial curve, with the starting point of the path as the origin and the actual curve length along the path as the position coordinate. This ensures that every point on the path has unique coordinates; each surface illumination anomaly sub-region is projected onto the path along a direction perpendicular to the path to obtain the interval occupied by that sub-region on the path. subscript Indicates the first A sub-region with abnormal surface illumination For the first The projected location of the center of each sub-region on the path (referred to as the "projection center") For the first The half-width of the projection of each sub-region along the path direction; Accordingly, determining the timing of illumination intensity changes during the movement of the machine vision light source based on the layout of all surface illumination anomaly sub-regions on the target object surface can specifically include the following steps: Real-time acquisition of the current coordinates of the machine vision light source on the path And for each sub-region Calculate the shortest arc distance from the current coordinates to the boundary of the sub-region. ; Based on the actual illumination intensity parameters of each sub-region, the illumination adjustment trigger distance for each sub-region is determined, and the global adjustment timing is determined by considering all sub-regions. The specific formula is as follows:

[0029] when When the machine vision light source reaches the point where the light intensity needs to be adjusted; The physical meaning, value range, and acquisition method of each parameter are as follows: The current arc length position of the machine vision light source, physically meaning the actual curve length the light source moves along the preset movement path from the starting point, is measured in units of length, and its value range is [value range missing]. ,in The total path length is obtained by real-time reading from the encoder of the motion control system. The number of encoder output pulses is calibrated and converted into arc length distance. Subscript Used to distinguish different surface illumination anomaly sub-regions, with a value range of [value range missing]. ,in This represents the total number of surface illumination anomalous sub-regions. For the first The projected center arc length of a sub-region with surface illumination anomalies is physically represented as the arc length corresponding to the center of that sub-region projected onto the path along a direction perpendicular to the path. Its dimension is length, and its value ranges from [value missing]. The specific acquisition method is as follows: perform region segmentation on the reference image and extract the first region. The center coordinates of each surface illumination anomaly sub-region are calculated. These center coordinates are projected onto the preset movement path along a direction perpendicular to the path, and the arc length corresponding to the projection point is calculated as... ; For the first The projected half-width of a surface illumination anomaly sub-region is physically defined as half the extent of that sub-region along the path direction, with dimensions of length and a value of [value missing]. The specific method for obtaining it is as follows: [The text abruptly ends here, likely due to an incomplete sentence or a format Each sub-region is projected onto a preset moving path along a direction perpendicular to the path, and the length of the projection interval of the sub-region on the path is obtained. Half of this length is taken as... ; Current arc length position To the The shortest arc length of a sub-region boundary, physically representing how far the light source needs to move along the path to enter the projection range of that sub-region, is measured in units of length and has a range of values. From the formula The calculation shows that, among which Indicates the current arc length position relative to the first arc. The absolute difference between the projection centers of each sub-region, minus The distance to the boundary is then obtained when the light source is within the projection range of the sub-region. ,at this time ; For the first The actual illumination intensity parameter of each surface illumination anomaly sub-region, its physical meaning is the optimal illumination intensity required for that sub-region to avoid overexposure or underexposure, with the dimension of light intensity and a value range of [value missing]. ,in This represents the maximum output illumination capability of the machine vision light source. The illuminance is defined as the light intensity under reference illumination conditions. Physically, it represents the standard illumination intensity set by the system under reference uniform illumination, with the dimension of luminous intensity and a value range of [value missing]. The intensity of light emitted from a target object can be obtained through calibration using a photoelectric sensor under reference illumination conditions. Specifically, when a reference uniform illumination is applied to the surface of the target object, a calibrated photoelectric sensor is used to measure the actual light intensity emitted by the light source onto the surface of the target object, and the average value of multiple measurements is taken as the standard illumination. ; For the first The dimensionless demand ratio of a sub-region, in physical terms, is the multiple of the required light intensity for that sub-region relative to the reference intensity, and its value ranges from [value range missing]. ,when The system will automatically Cut off as To keep this ratio no more than 1; For the first The lead factor for each sub-region, in physical terms, determines how far in advance the light source should begin adjusting the base length of the illumination. Its dimension is length, and its value is... Based on the geometric characteristics of this sub-region, it is predetermined. The specific method for obtaining it is as follows: for the first... Contour analysis was performed on a sub-region with surface illumination anomalies, and the area of ​​the sub-region was extracted. and mean curvature ,according to Calculation, where and For preset coefficients, The range of values ​​is , The range of values ​​is The specific values ​​were obtained through experimental calibration. The calibration method was as follows: test images were acquired under different lead settings, and the coefficient combination that made the light transition smoothest was selected. For the first The activation distance for lighting adjustment in each sub-region, in physical terms, is the current arc length position. Next, regarding the first The remaining distance threshold for triggering illumination adjustment, calculated for each sub-region, is measured in length and has a range of values. From the formula The calculation shows that this value represents the distance the light source still needs to move, and when the value is zero, it indicates that the critical point where adjustment needs to begin has been reached. The activation distance for global illumination adjustment is physically determined by considering all abnormal surface illumination sub-regions and determining the current arc length position. The adjustment distance threshold to be triggered below, with the unit being length, and the value range is [missing information]. From the formula The calculation shows that the minimum value among all sub-region start distances is taken, and when this value is zero, it is determined to be the time to adjust. The above technical solution establishes a path arc length coordinate system, projects each surface illumination anomaly sub-region onto the path, calculates a personalized start-up distance based on the actual illumination requirements of each sub-region, and then takes the minimum value of the start-up distances of all sub-regions as the global adjustment timing. Its beneficial effects are: it enables the light source to initiate illumination adjustments to multiple anomaly sub-regions in advance and smoothly during movement, avoiding sudden changes in illumination or response lag caused by different intensity requirements, thus improving the stability of illumination transitions and imaging continuity.

[0030] In another optional embodiment, determining the target value of the light intensity change during the movement of the machine vision light source based on the actual lighting parameters, and adjusting the light intensity change pattern of the target object during the movement of the machine vision light source based on the timing of the light intensity change and the target value of the light intensity change, can be implemented as follows: During the movement of the light source, the current arc length position of the light source is acquired in real time. and current movement speed ; Based on the difference in light intensity between each sub-region, the current position of the light source, and its current movement speed, the target light intensity value that the light source should output is determined using the following formula:

[0031] An upper limit constraint is imposed on the calculated target value of output illumination intensity:

[0032] When the aforementioned adjustment opportunity arrives, control the machine vision light source to output the target value; The physical meaning, value range, and acquisition method of each parameter are as follows: This represents the total number of surface illumination anomalous sub-regions. Physically, it refers to the number of sub-regions on the target object's surface that are identified as having illumination anomalous characteristics. The value is a positive integer. ; The light intensity under the reference lighting conditions; Subscript Used to distinguish different surface illumination anomaly sub-regions, with a value range of [value range missing]. ; For the first The difference in light intensity between sub-regions physically represents the deviation between the required light intensity of that sub-region and the reference light intensity. Its dimension is light intensity, and it can be positive or negative. Indicates the first Individual areas need increased light intensity. (This indicates a need to reduce light intensity), the specific method for obtaining this information is as follows: ,in For the first The actual light intensity parameters of each sub-region The light intensity under the reference lighting conditions; The current arc length position of the light source, physically meaning the actual curve length the light source moves along the preset path from the starting point, is measured in units of length and has a range of values ​​of [missing value]. ,in The total path length is obtained by real-time reading from the encoder of the motion control system. The number of encoder output pulses is calibrated and converted into arc length distance. For the first The projection center arc length of a sub-region, in physical terms, is the arc length corresponding to the projection of the center of that sub-region onto the path along a direction perpendicular to the path. Its dimension is length, and its value range is [value missing]. The specific acquisition method is as follows: perform region segmentation on the reference image and extract the first region. The center coordinates of each surface illumination anomaly sub-region are calculated. These center coordinates are projected onto the preset movement path along a direction perpendicular to the path, and the arc length corresponding to the projection point is calculated as... ; For the first The illumination influence range coefficient of a sub-region, in physical terms, determines the characteristic length by which the illumination compensation of that sub-region attenuates as the light source moves away. Its dimension is length, and its value ranges from [value missing]. The specific method for obtaining it is as follows: extract the first... The spatial distribution of high-frequency components of pixel contours within each sub-region is calculated. Taking the center of the sub-region as the origin, the attenuation curve of the high-frequency component density is calculated along the path direction, and the high-frequency component density is taken as the value at the center. Distance at time as The value of ; The Gaussian weighting function, in its physical sense, means that when the light source is located at an arc length position... At that time, the first The illumination compensation contribution weight of each sub-region is dimensionless and its value ranges from [value range missing]. Calculated by the formula, when The time weight is 1, when As the weight increases, the weight decays in the form of a Gaussian function. The current moving speed of the light source is, in physical terms, the instantaneous rate at which the light source moves along a preset path. Its dimension is velocity, and its value ranges from [value missing]. The specific method for obtaining it is: by determining the arc length position. The time derivative can be obtained, or the speed can be read in real time by the speed sensor in the motion control system; The standard moving speed, in physical terms, is the reference moving speed set by the system, with the dimension of velocity, and its value is... The specific acquisition method is as follows: pre-calibration is performed based on imaging quality and light source response characteristics. The calibration method is as follows: calibrated images are acquired at different moving speeds, and the speed that produces the highest image clarity is selected as the calibration speed. ; The velocity influence factor is a dimensionless coefficient that controls the sensitivity of light intensity output to changes in velocity. Its dimensionless value ranges from [value missing]. The specific method for obtaining the data is as follows: it is obtained through experimental calibration. The calibration method is to acquire calibration images at different movement speeds and adjust... To optimize imaging quality evaluation indicators (such as contrast and signal-to-noise ratio), or to determine them through system identification methods; The speed correction factor, physically speaking, is the overall scaling factor of the output light intensity when the moving speed of the light source deviates from the standard speed. It is dimensionless and is obtained through... Ensure this factor is non-negative to avoid negative light intensity output. Indicates the relative deviation of speed, when Time factor greater than 1 (increased light intensity compensation for high-speed movement), when Time factor less than 1 (reduce light intensity to avoid low-speed overexposure); The maximum output illumination capability of the light source, its physical meaning is the maximum illumination intensity that the machine vision light source can output, the dimension of which is light intensity, and it is an inherent parameter of the light source; This represents the target illuminance value that the light source should output at its current arc length position. Its physical meaning is: when the light source moves to the desired arc length position... At that time, the system calculates the location based on the projection center of each abnormal sub-region. Spatial relationships (through Gaussian weights) (reflection), and current movement speed (Reflected through a speed correction factor), the optimal output light intensity is calculated comprehensively. Therefore, yes The function changes dynamically as the light source moves along the path: when the light source approaches the projection center of a certain anomalous sub-region... At that time, the compensation amount of this sub-region The light is superimposed with higher weights to make the output light intensity approach the actual illumination intensity required for that sub-region; as the light source moves away, the compensation gradually decreases, and the output light intensity returns to the reference illumination intensity. Nearby. Meanwhile, Also depends on movement speed The overall scale adjusts to changes in speed, achieving adaptive speed compensation. This target value is constrained by an upper limit. Ensure that the output capacity of the light source is not exceeded.

[0033] The above technical solution employs a Gaussian weighting function to spatially superimpose the illumination compensation amounts for each sub-region, enabling a smooth transition in the output light intensity of the light source as its current position changes with the distance from the projection center of each sub-region, thus achieving spatially adaptive illumination compensation. Simultaneously, a speed correction factor is introduced, causing the output light intensity to dynamically scale with changes in the light source's movement speed, compensating for the impact of speed variations on exposure time, and achieving speed-adaptive dynamic response. The combination of these two methods ensures that the light source can accurately and smoothly meet the illumination needs of each abnormal sub-region during movement, preventing structural details from being obscured and improving the accuracy and reliability of machine vision recognition.

[0034] Please see Figure 2 As shown, an embodiment of this application provides an adaptive illumination intensity correction system for a machine vision light source. This adaptive illumination intensity correction system for a machine vision light source includes: The reference image processing module is used to acquire a reference image of the target object under reference lighting conditions and extract the pixel frequency domain distribution features of the reference image. The illumination anomaly calibration module is used to calibrate the surface illumination anomaly sub-regions of the target object based on the pixel frequency domain distribution characteristics. The adaptive lighting determination module is used to determine the adaptive lighting state of the surface lighting abnormal sub-region based on the lighting energy distribution of the surface lighting abnormal sub-region. The illumination parameter determination module is used to compare the baseline illumination conditions and the adaptive illumination state to determine the actual illumination parameters of the surface illumination anomaly sub-regions. The illumination mode adjustment module is used to adjust the illumination intensity change mode of the target object during the movement of the machine vision light source, based on the layout and actual illumination parameters of all surface illumination anomaly sub-regions.

[0035] The beneficial effects of the above embodiments are that the machine vision light source illumination intensity adaptive correction system extracts the pixel frequency domain distribution under reference illumination, calibrates abnormal sub-regions where the reference illumination cannot ensure the acquisition of structural detail information, and implements adaptive intensity variation illumination for all abnormal sub-regions to avoid obscuring structural details and improve the accuracy of machine vision recognition.

[0036] In another embodiment, the reference image processing module is used to acquire a reference image of the target object under reference illumination conditions, and extract the pixel frequency domain distribution features of the reference image, including: Acquire at least two reference images of the target object under reference uniform illumination in at least two directions; wherein the incident angle of the light rays of the reference uniform illumination relative to the surface of the target object satisfies a preset angle range, and the reference uniform illumination in at least two directions is applied alternately to the surface of the target object in time; The pixel frequency domain transformation of the reference image is performed to obtain the pixel contour frequency domain components of the reference image, and the average proportion of the pixel contour high-frequency components of each grid sub-region under all reference images is extracted. The illumination anomaly calibration module is used to calibrate the surface illumination anomaly sub-regions of a target object based on pixel frequency domain distribution characteristics, including: A threshold comparison is performed on the average proportion of high-frequency components of the pixel contour in each grid sub-region to determine whether the grid sub-region belongs to the abnormal surface illumination sub-region.

[0037] In another embodiment, the adaptive illumination determination module is used to determine the adaptive illumination state of the surface illumination anomalous sub-region based on the illumination energy distribution of the surface illumination anomalous sub-region, including: Frequency domain analysis is performed on the global pixel brightness of the surface illumination anomalous sub-region to obtain the frequency domain distribution of illumination energy; based on the frequency domain distribution of illumination energy, the overexposure degree of the surface illumination anomalous sub-region is determined, thereby obtaining the appropriate illumination intensity; The illumination parameter determination module is used to compare baseline illumination conditions with adapted illumination states to determine the actual illumination parameters of the surface illumination anomaly sub-regions, including: By comparing the baseline illumination intensity and the adapted illumination intensity obtained under the baseline illumination conditions in the surface illumination anomaly sub-region, the illumination intensity difference is obtained; based on the illumination intensity difference, the actual illumination intensity parameters of the surface illumination anomaly sub-region are determined.

[0038] In another embodiment, the illumination mode adjustment module is used to adjust the illumination intensity change mode of the target object during the movement of the machine vision light source based on the layout and actual illumination parameters of all surface illumination anomaly sub-regions, including: Based on the layout of all surface illumination anomaly sub-regions on the target object surface, determine the timing of illumination intensity changes during machine vision light source movement; Based on the actual illumination parameters, determine the target value of illumination intensity change during the movement of the machine vision light source; Adjust the illumination intensity change pattern of the target object during the movement of the machine vision light source based on the timing of the illumination intensity change and the target value of the illumination intensity change.

[0039] The illumination intensity adaptive correction system of the machine vision light source of the present invention has the same operation and effect as the above-mentioned illumination intensity adaptive correction method of the machine vision light source, and the illumination intensity adaptive correction system of the machine vision light source will not be described again here.

[0040] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. This disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims. Thus, if these modifications and variations of the invention fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A method for adaptive correction of illumination intensity of a machine vision light source, characterized in that, include: Obtain a reference image of the target object under reference lighting conditions, and extract the pixel frequency domain distribution features of the reference image; Based on the pixel frequency domain distribution characteristics, the surface illumination anomalous sub-regions of the target object are identified; Based on the light energy distribution of the surface illumination anomaly sub-region, determine the appropriate illumination state for the surface illumination anomaly sub-region; compare the reference illumination conditions and the appropriate illumination state to determine the actual illumination parameters of the surface illumination anomaly sub-region. Based on the layout of all surface lighting anomaly sub-regions and the actual lighting parameters, adjust the lighting intensity variation pattern of the target object during the movement of the machine vision light source.

2. The adaptive correction method for illumination intensity of machine vision light source as described in claim 1, characterized in that: Obtain a reference image of the target object under reference lighting conditions, and extract the pixel frequency domain distribution features of the reference image; Based on the pixel frequency domain distribution characteristics, the surface illumination anomalous sub-regions of the target object are identified, including: Acquire at least two reference images of a target object under reference uniform illumination along at least two directions; wherein the incident angle of the light rays of the reference uniform illumination relative to the surface of the target object satisfies a preset angle range, and the reference uniform illumination along at least two directions is applied alternately to the surface of the target object in time; The reference image is subjected to pixel frequency domain transformation to obtain the pixel contour frequency domain components of the reference image, and the average proportion of the pixel contour high-frequency components of each grid sub-region under all reference images is extracted. A threshold comparison is performed on the average proportion of high-frequency components of the pixel contour in each grid sub-region to determine whether the grid sub-region belongs to a surface illumination anomalous sub-region.

3. The adaptive correction method for illumination intensity of machine vision light source as described in claim 1, characterized in that: Based on the light energy distribution of the surface illumination anomaly sub-region, determine the appropriate illumination state for the surface illumination anomaly sub-region; By comparing the baseline illumination conditions and the adapted illumination state, the actual illumination parameters of the surface illumination anomaly sub-region are determined, including: Frequency domain analysis is performed on the global pixel brightness of the surface illumination anomaly sub-region to obtain the frequency domain distribution of illumination energy; based on the frequency domain distribution of illumination energy, the overexposure degree of the surface illumination anomaly sub-region is determined, thereby obtaining the appropriate illumination intensity; By comparing the reference illumination intensity and the adapted illumination intensity obtained by the surface illumination anomaly sub-region under the reference illumination conditions, an illumination intensity difference is obtained; based on the illumination intensity difference, the actual illumination intensity parameters of the surface illumination anomaly sub-region are determined.

4. The adaptive correction method for illumination intensity of machine vision light source as described in claim 1, characterized in that: Based on the layout of all surface illumination anomaly sub-regions and the actual illumination parameters, adjust the illumination intensity variation pattern of the target object during the movement of the machine vision light source, including: Based on the layout of all surface illumination anomaly sub-regions on the target object surface, determine the timing of illumination intensity changes during machine vision light source movement; Based on the actual illumination parameters, determine the target value of illumination intensity change during the movement of the machine vision light source; Based on the timing of the light intensity change and the target value of the light intensity change, the light intensity change pattern of the target object during the movement of the machine vision light source is adjusted.

5. An adaptive illumination intensity correction system for machine vision light sources, characterized in that, include: The reference image processing module is used to acquire a reference image of the target object under reference lighting conditions and extract the pixel frequency domain distribution features of the reference image; An illumination anomaly calibration module is used to calibrate the surface illumination anomaly sub-region of the target object based on the pixel frequency domain distribution characteristics; An adaptive illumination determination module is used to determine the adaptive illumination state of the abnormal surface illumination sub-region based on the illumination energy distribution of the abnormal surface illumination sub-region. The illumination parameter determination module is used to compare the reference illumination conditions and the adapted illumination state to determine the actual illumination parameters of the surface illumination anomaly sub-region. The illumination mode adjustment module is used to adjust the illumination intensity change mode of the target object during the movement of the machine vision light source based on the layout of all surface illumination abnormal sub-regions and the actual illumination parameters.

6. The machine vision light source illumination intensity adaptive correction system as described in claim 5, characterized in that: The reference image processing module is used to acquire a reference image of the target object under reference lighting conditions, and extract the pixel frequency domain distribution features of the reference image, including: Acquire at least two reference images of a target object under reference uniform illumination along at least two directions; wherein the incident angle of the light rays of the reference uniform illumination relative to the surface of the target object satisfies a preset angle range, and the reference uniform illumination along at least two directions is applied alternately to the surface of the target object in time; The reference image is subjected to pixel frequency domain transformation to obtain the pixel contour frequency domain components of the reference image, and the average proportion of the pixel contour high-frequency components of each grid sub-region under all reference images is extracted. The illumination anomaly calibration module is used to calibrate the surface illumination anomaly sub-regions of the target object based on the pixel frequency domain distribution characteristics, including: A threshold comparison is performed on the average proportion of high-frequency components of the pixel contour in each grid sub-region to determine whether the grid sub-region belongs to a surface illumination anomalous sub-region.

7. The machine vision light source illumination intensity adaptive correction system as described in claim 5, characterized in that: The adaptive illumination determination module is used to determine the adaptive illumination state of the abnormal surface illumination sub-region based on the illumination energy distribution of the abnormal surface illumination sub-region, including: Frequency domain analysis is performed on the global pixel brightness of the surface illumination anomaly sub-region to obtain the frequency domain distribution of illumination energy; based on the frequency domain distribution of illumination energy, the overexposure degree of the surface illumination anomaly sub-region is determined, thereby obtaining the appropriate illumination intensity; The illumination parameter determination module is used to compare the baseline illumination conditions and the adapted illumination state to determine the actual illumination parameters of the surface illumination anomaly sub-region, including: By comparing the reference illumination intensity and the adapted illumination intensity obtained by the surface illumination anomaly sub-region under the reference illumination conditions, an illumination intensity difference is obtained; based on the illumination intensity difference, the actual illumination intensity parameters of the surface illumination anomaly sub-region are determined.

8. The adaptive illumination intensity correction system for machine vision light sources as described in claim 5, characterized in that: The illumination mode adjustment module is used to adjust the illumination intensity change mode of the target object during the movement of the machine vision light source based on the layout of all surface illumination anomaly sub-regions and the actual illumination parameters, including: Based on the layout of all surface illumination anomaly sub-regions on the target object surface, determine the timing of illumination intensity changes during machine vision light source movement; Based on the actual illumination parameters, determine the target value of illumination intensity change during the movement of the machine vision light source; Based on the timing of the light intensity change and the target value of the light intensity change, the light intensity change pattern of the target object during the movement of the machine vision light source is adjusted.

Images