High-speed adaptive image acquisition system and method based on camera trigger light source

Abstract

The invention discloses a high-speed adaptive image acquisition system and method based on a camera trigger light source, and the system comprises an image acquisition unit which is fixedly disposed on a carrier, and is used for collecting an image of a to-be-detected road surface, and outputting an exposure indication signal; the trigger light source unit comprises a light source with adjustable pulse width and time delay and a driving module of the light source; the motion information acquisition unit is fixedly mounted on the carrier and is used for acquiring motion speed information of the image acquisition unit in real time; and the synchronous control unit is used for receiving the motion speed information and determining the exposure time of the image acquisition unit based on the motion blur constraint and the motion speed information. The light source response delay is actively compensated by performing time sequence calibration on the exposure time of the image acquisition unit and the light source highlight window, so that stable and sufficient light energy is ensured to accurately cover the whole light sensing process of the sensor within extremely short exposure time.

Description

Technical Field

[0001] This invention relates to the field of road vision inspection technology, specifically to a high-speed adaptive image acquisition system and method based on a camera-triggered light source. Background Technology

[0002] In the field of engineering inspection, manual visual inspection has historically been the primary method for detecting cosmetic defects in basic transportation infrastructure such as roads, bridges, and tunnels. This method relies heavily on repetitive manual labor, involving the observation and recording of each structure to generate inspection results and reports. During inspection, defects are primarily recorded in writing and supplemented by photographs. However, differences in the habits of the photographers and variations in the descriptions of defects by different inspectors lead to discrepancies in the reports. In recent years, with the rise of automated inspection, automated inspection has become the main method for detecting cosmetic defects.

[0003] In public transportation inspections, such as road inspections, to ensure smooth traffic flow, it's impossible to use closed-off methods. Road inspection vehicles need to maintain high-speed movement to collect road defect images. To avoid motion blur, microsecond-level short exposures are typically used. Simultaneously, due to unstable ambient light, high-brightness stroboscopic light sources are required for supplemental illumination. Common road defect images are generally acquired using grayscale images. Grayscale images are characterized by only brightness variations, clear structure, texture, and edge information, and smaller data volume compared to color images. They have good recognition effects on common asphalt pavement defects such as rutting, loosening, potholes, network cracks, and longitudinal and transverse cracks, as well as concrete pavement defects such as cracks, broken slabs, peeling, and chipped corners. However, they are less effective at recognizing asphalt pavement defects such as oil stains, water stains, bleeding, material segregation, and patches, and concrete pavement defects such as water seepage, water spots, efflorescence, pollution, and color differences in repaired areas, causing these areas to be confused with ordinary shadows and road color variations. Therefore, depending on different road conditions and storage requirements, the use of both color and grayscale cameras should be considered in the inspection process.

[0004] However, existing high-speed adaptive image acquisition systems still have the following problems:

[0005] 1. The delay problem of camera triggering light source: The camera usually outputs a trigger signal to control the light source. There is a fixed or temperature-related response delay from the time the light source receives the trigger signal to the time it actually emits light. Under short exposure conditions, if this delay is not compensated, the camera exposure time window may be misaligned with the high brightness segment of the light source, resulting in insufficient lighting, dark images, or high noise.

[0006] 2. Exposure parameters and light source timing are configured separately. Traditional systems typically set exposure time, gain, light source pulse width, and trigger delay separately, lacking a unified closed-loop control with image quality as feedback, and relying on repeated manual parameter adjustments.

[0007] 3. The brightness of grayscale mode and color mode is inconsistent. Under the same light source and scene, the grayscale image looks normal in brightness, but the color image is obviously dark. When changing cameras, it is necessary to manually readjust the exposure, gain and white balance, which is inefficient and difficult to ensure consistency. Summary of the Invention

[0008] The purpose of this invention is to provide a high-speed adaptive image acquisition system and method based on a camera-triggered light source, so as to at least solve the current problems.

[0009] To address the aforementioned technical problems, in a first aspect, the present invention provides a high-speed adaptive image acquisition system based on a camera-triggered light source, comprising:

[0010] The image acquisition unit is fixedly mounted on the carrier and is used to acquire images of the road surface to be tested and output exposure indication signals;

[0011] The trigger light source unit includes an adjustable pulse width and delay light source and its driving module;

[0012] The motion information acquisition unit is fixedly installed on the carrier and is used to acquire the motion speed information of the image acquisition unit in real time.

[0013] A synchronization control unit is used to receive the motion speed information and determine the exposure time of the image acquisition unit based on motion fuzz constraints and the motion speed information.

[0014] Furthermore, methods for determining exposure time include:

[0015] Based on the motion speed v of the tested carrier, the imaging resolution R of the image acquisition unit, and the maximum allowable number of blurred pixels M, the upper limit of the exposure time T is calculated. MAX The exposure time of the image acquisition unit is determined by taking into account the frame rate constraint and the value ≤ M*R / v.

[0016] Furthermore, timing calibration methods include:

[0017] The actual time position and width of the exposure time are obtained based on the exposure indication signal output by the image acquisition unit;

[0018] Trigger delay T under different light sources delay Pulse width T L Combine images of the signboard and calculate the average gray level of the captured images;

[0019] Select the parameter combination (T) that makes the average gray level close to the maximum value before saturation and minimizes fluctuation. * delay T * L The best matching point is selected and saved as the calibration result.

[0020] Furthermore, the image acquisition unit includes:

[0021] A grayscale camera is used to acquire grayscale images of the road surface under test in typical scenarios where there are no obvious road surface defects.

[0022] A color camera is used to acquire color images of the road surface under test in atypical scenarios where there are obvious defects in the road surface.

[0023] Furthermore, the synchronization control unit also includes uniformly calibrating the brightness of the grayscale camera and the color camera, specifically including:

[0024] Under the same scene and the same light source parameters, multiple grayscale images and multiple color images were acquired using a grayscale camera and a color camera, respectively.

[0025] Calculate the average gray value L of multiple grayscale images respectively. mono and the average brightness value L of multiple color images color ;

[0026] With average gray value L mono Using brightness as a benchmark, the exposure time and gain of the color camera are iteratively adjusted to achieve an average brightness value L. color With average gray value L mono The difference between them does not exceed the preset threshold ΔL th .

[0027] Furthermore, during system operation, image brightness is statistically analyzed in real time, and the exposure time and gain of the color camera are dynamically adjusted based on the statistical results.

[0028] Furthermore, the methods for dynamically adjusting the exposure time and gain of a color camera include:

[0029] Set the target brightness range [L] _min L _max ];

[0030] If the average brightness L color Below L _min If the exposure time reaches the maximum exposure time T, then the exposure time will be increased first. MA Then, increase the gain;

[0031] If the average brightness L color Higher than L _max If necessary, reduce the gain first, and then reduce the exposure time if necessary.

[0032] Furthermore, the carrier is a vehicle.

[0033] Secondly, the present invention provides a high-speed adaptive image acquisition method based on a camera-triggered light source. This method is implemented based on the system provided in the first aspect above and includes the following steps:

[0034] S1: The image acquisition unit acquires an image of the road surface to be tested and outputs an exposure indication signal; the motion information acquisition unit acquires the motion speed information of the image acquisition unit.

[0035] S2: Based on motion fuzzy constraints and motion speed information, determine the exposure time of the image acquisition unit; then perform timing calibration on the exposure time of the image acquisition unit and the highlight window of the light source, and control the drive module to ensure that the highlight period of the light source completely covers the exposure time of the image acquisition unit on the time axis.

[0036] The beneficial effects of this invention are as follows: by using motion speed as the core input, the exposure time that does not produce blur is dynamically determined, thus ensuring the basic clarity of the image from the source; by performing time-series calibration of the exposure time of the image acquisition unit and the high-brightness window of the light source, the light source response delay is actively compensated, ensuring that stable and sufficient light energy accurately covers the entire sensor photosensitive process within an extremely short exposure time, thus solving the problems of dark images and high noise caused by short exposure, thereby obtaining images with high definition and high signal-to-noise ratio under high-speed conditions. Attached Figure Description

[0037] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, use the same reference numerals to denote the same or similar parts. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0038] Figure 1 This is a system block diagram of one embodiment of the present invention. Detailed Implementation

[0039] like Figure 1 The high-speed adaptive image acquisition system based on a camera-triggered light source, as shown, includes:

[0040] The image acquisition unit is fixedly mounted on the carrier and is used to acquire images of the road surface to be tested and output exposure indication signals;

[0041] The trigger light source unit includes an adjustable pulse width and delay light source and its driving module; the light source includes a set of high-brightness LED area array lights, the driving module is a matching constant current driving power supply, supports TTL trigger input, the light source pulse width is adjustable, and it has a fast rise time and a clear response delay; the light source brightness can be adjusted by both driving current and pulse width.

[0042] The motion information acquisition unit is fixedly mounted on the carrier and is used to acquire the motion speed information of the image acquisition unit in real time. The motion information acquisition unit uses an encoder and a tire to form a coaxial system to acquire the motion speed V (m / s) of the target relative to the camera in real time.

[0043] The synchronization control unit receives motion speed information and determines the exposure time of the image acquisition unit based on motion fuzz constraints and motion speed information. Then, it performs timing calibration on the exposure time of the image acquisition unit and the high-brightness window of the light source. Based on the calibration results, it drives the light source module to ensure that the high-brightness period of the light source completely covers the exposure time of the image acquisition unit on the time axis. The synchronization control unit uses a programmable logic controller (PLC) for synchronization control. It receives motion speed information and controls the camera trigger, i.e., external camera trigger output (triggering the camera's exposure start). When the camera receives the trigger signal and begins exposure, the camera trigger electrical signal excites the light source. After receiving the trigger signal, the light source driving module determines the exposure time based on the light source trigger delay T. * delay Pulse width T * L Output light source drive signal.

[0044] This invention uses motion speed as the core input to dynamically determine the exposure time that does not produce blur, thus ensuring the basic clarity of the image from the source. By timing the exposure time of the image acquisition unit and the high-brightness window of the light source, it actively compensates for the light source response delay, ensuring that stable and sufficient light energy accurately covers the entire sensor photosensitive process within an extremely short exposure time. This solves the problems of dark images and high noise caused by short exposure, thereby obtaining images with high definition and high signal-to-noise ratio under high-speed conditions.

[0045] According to one embodiment of this application, the method for determining the exposure time includes:

[0046] Based on the motion speed v of the tested carrier, the imaging resolution R of the image acquisition unit, and the maximum allowable number of blurred pixels M, the upper limit of the exposure time T is calculated. MAX ≤M*R / v, and taking into account the frame rate constraint, determine the exposure time of the image acquisition unit; where v is the relative motion speed of the camera, in m / s; R is the ground resolution or object resolution, in m / pixel; M is the maximum allowable blur length M, usually taken as 1 or 2.

[0047] Frame rate constraints include: In global shutter high-speed cameras, the exposure time must also be less than the frame period.

[0048] Frame rate: f

[0049] Frame rate: T frame =1 / f

[0050] T≤T frame =1 / f

[0051] Typically, in most industrial cameras, the drive requires the exposure time to not exceed a certain duty cycle:

[0052] T≤K·T frame =1 / f (0 < k < 1)

[0053] In the formula, K is generally 0.9 or less to allow for processing time.

[0054] For example, if the camera's field of view for the road surface is 1.4m, and a vehicle travels at 40km / h (approximately 11.11m / s), the distance the vehicle travels in 1 second is approximately 11.11m. To ensure that at least one frame is captured for each segment of the field of view along the vehicle's direction of travel, the required minimum sampling frame rate can be obtained as follows:

[0055] f need =v / L=11.11 / 1.4=7.94fps

[0056] That is, when the frame rate is not less than about 8fps, one frame of vehicle image can be obtained every 1.4m on the road surface. In this embodiment, the camera's working frame rate is set to 15fps, which is greater than the minimum frame rate requirement mentioned above, thus providing a certain margin in terms of road surface coverage sampling.

[0057] T frame =1 / 15=66.7ms

[0058] Considering that the camera also needs to complete charge readout and data transmission, the maximum duty cycle of the exposure time within the frame period is preferably set to 0.8, so the upper limit of the exposure time given by the frame rate is approximately 50-55ms. On the other hand, since the vehicle continues to move along the road direction during the exposure time, in order to ensure that the displacement of the same target on the imaging plane does not exceed M pixels, motion blur constraints are introduced in this embodiment:

[0059] T≤MR / v

[0060] Where R is the object space resolution, and in this embodiment, a field of view of 1.4m corresponds to 5120 pixels, therefore:

[0061] R = 1.4 / 5120 = 0.000273 m / pixel

[0062] When the maximum allowable blur amount M = 1 pixel, the result is:

[0063] T≤1*0.000273 / 11.1=24.6 microseconds

[0064] Therefore, the final upper limit of the exposure time can be obtained as 24.6 microseconds. Thus, under the conditions of a vehicle speed of 40km / h and a field of view of 1.4m, in order to balance road surface coverage sampling and image clarity, the upper limit of the exposure time in this embodiment is determined to be about 20 to 25µs. Preferably, the working exposure time of the camera is set in the range of 15 to 20µs.

[0065] According to one embodiment of this application, the timing calibration method includes:

[0066] The actual time position and width of the exposure time are obtained based on the exposure indication signal output by the image acquisition unit. The time position refers to the start and end times of the exposure window on the time axis, that is, the moment the exposure begins and the moment it ends. It is usually expressed as a delay relative to a certain time reference (such as the rising or falling edge of the trigger signal; in this camera, it is the falling edge). Time position = exposure start time (start) and exposure end time (end), width = end - start, i.e., exposure time. Therefore, the purpose of calibration is to ensure that the highlight window of the light source completely covers this exposure window in time.

[0067] Trigger delay T under different light sources delay Pulse width T L The image is photographed in combination with other images (uniform white board or diffuse reflection board), and the average gray level of the captured image is calculated. Both of these parameters directly affect the brightness of the captured image. After determining a fixed exposure time, the light source trigger delay and pulse width are changed one by one. The average brightness of the image under each combination is calculated. The combination with the highest and most stable brightness is selected to find the optimal light source trigger parameters so that the light source is fully lit during the exposure period (similar to the controlled variable method).

[0068] Select the parameter combination (T) that makes the average gray level close to the maximum value before saturation and minimizes fluctuation. * delay T * L The best matching point is selected and saved as the calibration result.

[0069] In "Camera-triggered light source" mode, the camera outputs a control signal synchronized with the exposure time during each frame exposure. The duration of the high level of this signal corresponds to the camera's actual exposure time (the actual exposure time of the camera usually refers to the real-time image seen through the camera's viewfinder or display screen, which reflects the exposure effect under the current settings). This invention calibrates the relative timing between the exposure time and the light source's highlight window in one step, obtaining the light source response delay and optimal trigger delay parameters. This ensures that the highlight period of the light source completely covers the camera's exposure time on the time axis, thereby achieving stable, sufficient, and uniform illumination even with short exposure times. The actual position and width of the exposure time window are obtained through the exposure signal provided by the camera. The camera outputs a hardware signal indicating the effective exposure range, the duration of which corresponds to the actual exposure time of the current frame. The rising and falling edges of the signal correspond to the start and end times of the exposure, respectively. In this invention, the light source requires the camera's falling edge to trigger.

[0070] According to one embodiment of this application, the image acquisition unit includes:

[0071] A grayscale camera is used to acquire grayscale images of the road surface under test in typical scenarios where there are no obvious road surface defects.

[0072] A color camera is used to acquire color images of the road surface under test in atypical scenarios where there are obvious defects in the road surface.

[0073] Grayscale cameras, with their high sensitivity and high resolution, accurately capture structural defects such as cracks and textures; color cameras, on the other hand, can effectively identify color-related defects such as oil stains and color differences. This hardware configuration provides the physical possibility for comprehensive defect detection.

[0074] According to one embodiment of this application, the synchronization control unit further includes uniformly calibrating the brightness of the grayscale camera and the color camera, specifically including:

[0075] Under the same scene and with identical light source parameters, multiple grayscale images and multiple color images are acquired using a grayscale camera and a color camera, respectively. When acquiring the grayscale image of the road surface to be measured using the grayscale camera, the gain G of the grayscale camera is first determined. mono Based on the light source parameters and the exposure time determined in the preceding steps, 50 to 100 images from different locations are acquired, and the average gray value L of the images is calculated. mono ;

[0076] The color camera sets the initial exposure T while keeping the light source parameters constant. 0 color =T mono Gain G 0 color =G monoTurn off white balance, and then acquire a color image.

[0077] Calculate the average gray value L of multiple grayscale images respectively. mono and the average brightness value L of multiple color images color Average gray value L mono The average brightness value L is calculated directly from the grayscale values ​​of the grayscale image. color The calculation methods include:

[0078] The acquired color image is converted into a luminance channel, and the average luminance value is calculated. Lcolor ;

[0079] By iteratively adjusting the exposure time in color mode (not exceeding the aforementioned upper limit), overall gain, and white balance common gain, thus...

[0080] L color -L mono ≤ΔL th (Preset brightness error threshold)

[0081] Finally, the obtained parameter combination will be compared with the grayscale mode reference parameters (G). mono T mono Establish a mapping relationship and store it in the configuration file or the camera's internal non-volatile storage.

[0082] This embodiment establishes a brightness mapping relationship between the outputs of the two types of cameras by uniformly calibrating the brightness of grayscale and color modes. This provides a unified brightness benchmark for subsequent image processing algorithms, solves the problem of inconsistent brightness in grayscale and color modes for the same scene, and avoids the problem of having to readjust the algorithm due to switching acquisition modes.

[0083] According to one embodiment of this application, during the system operation phase, the image brightness is statistically analyzed in real time (e.g., in each acquisition or every N frames of acquisition, a region of interest (ROI) or the entire image is set and the brightness value is calculated), and the exposure time and gain of the color camera are dynamically adjusted based on the statistical results.

[0084] According to one embodiment of this application, a method for dynamically adjusting the exposure time and gain of a color camera includes:

[0085] Set the target brightness range [L] _min L _max ];

[0086] If the average brightness L color Below L _min If the exposure time reaches the maximum exposure time T, then the exposure time will be increased first. MA Then, appropriately increase the camera gain;

[0087] If the average brightness L color Higher than L _max If the image is still too bright, then reduce the camera gain first. If the image is still too bright, then reduce the exposure time to avoid saturation.

[0088] During system operation, adaptive exposure and light source control based on real-time brightness feedback are introduced to form a closed-loop control. This allows the system to automatically fine-tune parameters according to changes in ambient light and road surface reflectivity, stabilizing image brightness within the optimal range. This significantly reduces reliance on human experience and ensures the long-term stability and comparability of image quality acquired at different times and on different road sections.

[0089] According to one embodiment of this application, the carrier is a vehicle.

[0090] Secondly, the present invention provides a high-speed adaptive image acquisition method based on a camera-triggered light source. This method is implemented based on the system provided in the first aspect above and includes the following steps:

[0091] S1: The image acquisition unit acquires an image of the road surface to be tested and outputs an exposure indication signal; the motion information acquisition unit acquires the motion speed information of the image acquisition unit.

[0092] S2: Based on motion fuzzy constraints and motion speed information, determine the exposure time of the image acquisition unit; then perform timing calibration on the exposure time of the image acquisition unit and the highlight window of the light source, and control the drive module to ensure that the highlight period of the light source completely covers the exposure time of the image acquisition unit on the time axis.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A high-speed adaptive image acquisition system based on a camera-triggered light source, characterized in that, include: The image acquisition unit is fixedly mounted on the carrier and is used to acquire images of the road surface to be tested and output exposure indication signals; The trigger light source unit includes an adjustable pulse width and delay light source and its driving module; The motion information acquisition unit is fixedly installed on the carrier and is used to acquire the motion speed information of the image acquisition unit in real time. A synchronization control unit is used to receive the motion speed information and determine the exposure time of the image acquisition unit based on the motion fuzzy constraints and the motion speed information. Then, the exposure time of the image acquisition unit and the high-brightness window of the light source are time-series calibrated. Based on the calibration results, the driving module is controlled to drive the light source so that the high-brightness period of the light source completely covers the exposure time of the image acquisition unit on the time axis.

2. The high-speed adaptive image acquisition system based on a camera-triggered light source according to claim 1, characterized in that, The method for determining the exposure time includes: Based on the motion speed v of the tested carrier, the imaging resolution R of the image acquisition unit, and the maximum allowable number of blurred pixels M, the upper limit of the exposure time T is calculated. MAX Taking frame rate constraints into account, the exposure time of the image acquisition unit is determined; the formula for calculating the upper limit of the exposure time is: T MAX ≤M*R / v。 3. The high-speed adaptive image acquisition system based on a camera-triggered light source according to claim 2, characterized in that, The timing calibration method includes: The actual time position and width of the exposure time are obtained based on the exposure indication signal output by the image acquisition unit; Using the image acquisition unit to trigger delay T under different light sources delay Pulse width T L The signboard was photographed using a combination of methods, and the trigger delay T for different light sources was calculated. delay Pulse width T L Different average gray levels of images captured under combination; Select the parameter combination (T) that makes the average gray level close to the maximum value before saturation and minimizes fluctuation. * delay T * L The best matching point is selected and saved as the calibration result.

4. The high-speed adaptive image acquisition system based on a camera-triggered light source according to any one of claims 1-3, characterized in that, The image acquisition unit includes: A grayscale camera is used to acquire grayscale images of the road surface under test in typical scenarios where there are no obvious road surface defects. A color camera is used to acquire color images of the road surface under test in atypical scenarios where there are obvious defects in the road surface.

5. The high-speed adaptive image acquisition system based on a camera-triggered light source according to claim 4, characterized in that, The synchronization control unit also includes uniformly calibrating the brightness of the grayscale camera and the color camera, specifically including: Under the same scene and the same light source parameters, multiple grayscale images and multiple color images are acquired using the grayscale camera and the color camera, respectively. Calculate the average gray value L of multiple grayscale images respectively. mono and the average brightness value L of multiple color images color ; The average gray value L mono Using brightness as a reference, the exposure time and gain of the color camera are iteratively adjusted to achieve an average brightness value L. color With average gray value L mono The difference between them does not exceed the preset threshold ΔL th .

6. The high-speed adaptive image acquisition system based on a camera-triggered light source according to claim 5, characterized in that, During system operation, image brightness is statistically analyzed in real time, and the exposure time and gain of the color camera are dynamically adjusted based on the statistical results.

7. The high-speed adaptive image acquisition system based on a camera-triggered light source according to claim 6, characterized in that, The method for dynamically adjusting the exposure time and gain of the color camera includes: Set the target brightness range [L] _min L _max ]; If the average brightness L color Below L _min If the exposure time is increased, the exposure time will be increased first. If the exposure time reaches the upper limit T, the exposure time will be increased first. MA Then, increase the gain; If the average brightness L color Higher than L _max If necessary, reduce the gain first, and then reduce the exposure time if necessary.

8. The high-speed adaptive image acquisition system based on a camera-triggered light source according to claim 5, characterized in that, The carrier is a vehicle.

9. A high-speed adaptive image acquisition method based on a camera-triggered light source, characterized in that, This method is implemented based on the system described in any one of claims 1-8, and includes the following steps: S1: The image acquisition unit acquires an image of the road surface to be tested and outputs an exposure indication signal; the motion information acquisition unit acquires the motion speed information of the image acquisition unit. S2: Based on motion fuzzy constraints and the motion speed information, determine the exposure time of the image acquisition unit; then perform timing calibration on the exposure time of the image acquisition unit and the highlight window of the light source, and control the driving module to ensure that the highlight period of the light source completely covers the exposure time of the image acquisition unit on the time axis.

Images