KR20120000234A - The method of auto-exposure control for white light 3d scanner using an illuminometer - Google Patents

Abstract

PURPOSE: An automatic exposure control method of a white light three dimensional scanner using an illumination sensor is provided to improve reliability for restored three dimensional image by controlling a light emitting device. CONSTITUTION: An illumination sensor detects light from an external environment(2). A projector creates a pattern(4). A three-dimensional scanner projects the pattern(5). The three-dimensional scanner acquires images through a camera. The three-dimensional scanner analyzes a corresponding point error(8). The three-dimensional scanner performs a three-dimensional computation using a triangulation method(9).

Description

The method of auto-exposure control for white light 3D scanner using an illuminometer}

The present invention relates to a pattern scanning brightness and exposure time determination algorithm and a system configuration thereof for a three-dimensional scanner of white light type, more specifically, to automatically determine the optimal brightness and exposure time for changes in the ambient light to control the brightness of the projector. The present invention relates to an optimal brightness control and exposure time control method and system of a white light-based three-dimensional scanner capable of increasing reliability of a restored three-dimensional image by adjusting a PWM value and an exposure time of a camera.

The white light-based three-dimensional camera is composed of a projector which is a projection means of a pattern and a camera which is an acquisition means of a pattern image. In the projector, a pattern is irradiated to an object, an object to which the pattern is irradiated is acquired by a camera, and three-dimensional information is obtained through three-dimensional processing of the acquired image.

Laser scanners, which are conventional active sensors, are generally used because of their high accuracy, but they are not easily used due to the limitation and cost of the scanning speed. The stereo camera system is affected by the surrounding environment because it uses the binocular difference between two images as a feature point from the image. In particular, it has a disadvantage in that accurate three-dimensional data cannot be generated in an environment without a feature such as a wall.

On the other hand, the white light-based three-dimensional scanner uses the projected pattern as a feature point, and thus can actively acquire three-dimensionally. Projecting the pattern from the projector acquires the pattern image from the camera and calculates the pixel correspondence point between the camera and the projector. At this time, the correct pixel correspondence is determined only when the pattern of the image acquired by the camera is properly distinguished. Based on this correspondence, the object is reconstructed in three dimensions by performing calculation using the triangulation technique. White light-based three-dimensional scanner has the advantage of very high three-dimensional data accuracy compared to the stereo or TOF (time of flight) three-dimensional scanner, but also has the following disadvantages.

First, the 3D scanner based on white light can increase the 3D reconstruction reliability only by acquiring the exact correspondence between the projector and the camera. A pixel correspondence point is a position value of an image at a projector position with respect to an image of a camera, and is generally referred to as an 'address value' or a 'correspondence point'. The pixel correspondence points are constantly increased from the left to the right side of the image according to the codec used, and are mostly expressed in numbers. Only when the corresponding point is obtained correctly, the precision of the 3D reconstructed data obtained by the triangulation method is increased. The corresponding point can be obtained correctly only when the pattern image of the projector acquired by the camera is properly distinguished. However, a change in the intensity of the signal occurs according to the change of illumination, and thus it is difficult to distinguish the acquired pattern in a very bright environment, and it is difficult to obtain an accurate pattern due to light bleeding and mixing in a very dark environment.

Secondly, white light-based three-dimensional scanners require a projector to illuminate the pattern. The projector projects a pattern using a light source, and various light sources can be used depending on the purpose. Some light sources, such as halogen and Flash, are difficult to control light intensity, but some projectors use light sources that can control light intensity using PWM signals, such as LEDs, OLEDs, and OFDs, which are semiconductor light emitting devices. When the projector using the semiconductor light emitting device is used, the life of the light emitting device may be shortened due to the high power consumption and heat generation due to the current control driving method. In particular, semiconductor light emitting devices generate a maximum amount of light at the beginning of the initial current supply. If the light is continuously turned on, the projector cannot use the maximum amount of light. Therefore, since the light intensity of the pattern projected by the projector is not bright, it is difficult to distinguish the exact pattern because the quality of the pattern image acquired by the camera is not good.

Therefore, the present invention controls the optimal brightness of the projector light and the camera exposure according to the characteristics of the surrounding environment to be restored on the basis of the white light, thereby enabling accurate pixel correspondence to increase the reliability of the 3D result.

3 is performed in a 3D scanner based on white light by determining and adjusting an optimal projector pattern brightness level and a camera exposure level that can most appropriately determine a pattern reflected by an object according to a variety of changing 3D measurement environments. To provide the best way to improve the dimensional data precision and reliability.

By detecting the brightness of the 3D environment to be measured using the digital illuminance sensor, the exposure time of the camera and the brightness of the projector may be controlled based on the detected value.

If the white light three-dimensional scanner is developed through the present invention, the reliability of the three-dimensional image is high, it can be utilized in reverse engineering, medical, cultural property restoration. It can also be used as a handheld scanner because it does not consume much power by controlling the light emitting device so that the projector can be turned on only when you want to use it.

1 is a block diagram showing an exposure system configuration of a white light 3D scanner;
2 is a block diagram showing camera exposure and projector illumination control of a three dimensional scanner of the present invention;
3 is a block diagram showing a scanner exposure control through the correspondence point error determination in the present invention;
* Description of the symbols for the main parts of the drawings *
1: Start 3D scanner 2: Illumination detection by illuminance sensor
3: lighting unit control unit 4: pattern generator of the projector
5: camera projector synchronization and camera image acquisition
6: Counter for Counting Camera Image Acquisition
7: Camera-projector correspondence point calculation
8: correspondence point error analysis unit
9: Acquiring 3D distance based on the correspondence point
* Lighting control unit
31: Comparison with the illuminance detected by the illuminance sensor
32: Increase the projector light brightness
33: Use the projector brightness as it is
34: Reduces projector light brightness
35: reduce camera exposure time
36: Camera exposure time as it is
37: increase camera exposure time
Correspondence point error analysis unit
81: comparison with the threshold error value based on the acquired correspondence point
82: extraction of correspondence point error and duplicate value occurrence
83: 3D scanner retake

The present invention provides a structured light three-dimensional scanner system for restoring three-dimensional data by irradiating a pattern to a target object from a projector and a projection device and obtaining the pattern by an image capturing means such as a camera. Projector brightness control and camera exposure time determination and systems therefor.

In order to determine the optimal brightness and exposure in the system, the brightness and exposure time are controlled by comparing the illumination value input from the external illumination sensor with the reference illumination set by the user. In addition, after calculating the corresponding point of the projector and the camera, the error of each point is analyzed and the reflectance is measured, and the brightness and exposure time are automatically determined based on this. By using the projector as a projection device, the pattern irradiation time is driven for a limited time only in accordance with the exposure time of the camera, so as to obtain a good image quality by using the maximum amount of light generated at the beginning of the current supply. It is characterized by reducing power consumption.

In the white light 3D scanner, the brightness control and exposure control are performed according to the environmental conditions as follows.

First, using the illuminance sensor to detect the illuminance of the external environment and compare the internally set threshold illuminance: The illuminance sensor is mounted on the outside of the system to measure the illuminance of the external environment at the time of initial system operation or periodically. The measured illuminance value is compared with a threshold illuminance value set inside the system, and then the next step is executed according to the comparison value.

Second, when the detected illuminance is smaller than the minimum value of the critical illuminance (Min), the external environment is judged as a dark environment to reduce the PWM duty value of the projector to lower the pattern irradiation brightness and increase the exposure time of the camera: the detected illuminance value If it is smaller than the minimum value of the set threshold illuminance, it means that the external environment is dark and it is difficult to obtain an accurate 3D image. At this time, if the projector light is too bright, the pattern image acquired by the camera will blur. Therefore, the projector brightness should be lowered. Therefore, by reducing the PWM duty value of the projector to lower the brightness, the camera increases the exposure time to obtain a bright image to increase the three-dimensional reconstruction accuracy.

Third, if the detected illuminance is greater than the minimum value of the critical illuminance (Min), the external environment is determined to be a very bright environment, increasing the PWM duty value of the projector to increase the pattern irradiation brightness and reduce the exposure time of the camera: the detected illuminance If the value is larger than the minimum value of the set threshold illuminance, it means that the external environment is bright and it is difficult to obtain an accurate 3D image. In a bright environment, the distinction between the patterns projected by the projector is ambiguous. In this case, the brightness of the projector is increased so that the patterns can be accurately distinguished. Therefore, by increasing the PWM duty value of the projector to increase the brightness of the light, the camera reduces the exposure time to obtain a dark image to increase the three-dimensional reconstruction accuracy.

Fourth, when performing the 3D system operation for the first time while the detection illuminance falls within the range of the threshold illuminance value, when the pattern is generated by the projector and the light is projected to project the pattern, acquiring an image using a camera: the second B) When the detected illuminance value is in the range of the threshold illuminance value through the third step, pattern projection and image acquisition operations for 3D reconstruction are performed. When the pattern is created in the projector module, turn on the light to project the pattern. The projected pattern acquires an image using a camera. This operation is repeatedly performed until the images capable of 3D processing are collected.

Fifth, counting the position of the point of occurrence of the error on the basis of the calculated value of the corresponding point, and comparing and analyzing the threshold error count value set based on this: the camera and the camera for 3D image processing with the image obtained in the fourth step Calculate the correspondence between projectors. The number of occurrences of an error is counted at a corresponding point calculated in all the image pixels and compared with a set threshold error count value. If it is smaller than the threshold error count range, the final three-dimensional operation is performed based on the corresponding point, and if it is larger than the threshold error count range, the error pattern is analyzed.

Sixth, determining whether the error is caused by the shadow or the discontinuous plane according to the error type. When the shadow or the discontinuous plane occurs in the object or the environment, the error occurs because the pattern of the projector is not projected in the shadow area of the object or the environment. If there is a discontinuity of objects or if the pattern is not projected in the shadowed area, the error correspondence point is generally expressed as '0' because the correspondence point between the projector and the camera cannot be found. Corresponding points in space should increase constantly from left to right when expressed as a number, which increases constantly and is marked as '0' in some areas, and if it increases again, it means a discontinuity or shadow. The white light 3D technique can only restore 3D to the projected area, so 3D measurement is possible only by changing the position of measurement or the shape of the object to be measured. Therefore, in this case, the final three-dimensional operation is performed using only the measured corresponding point. However, if the correspondence point increases constantly but the correspondence point value is duplicated and the repetition occurs continuously, this is a case where the exact correspondence point cannot be extracted because the projector cannot accurately acquire the pattern. In this case, discard the acquired value of the corresponding point and go back to the first step to control the camera exposure and the projector brightness again to acquire a pattern image.

The final three-dimensional calculation is performed by using a triangulation method by using the optimum point of the projector light control and the corresponding point obtained through the camera exposure step. The result of the performed 3D image is high in flexibility with respect to a reconstruction environment and has high reliability.

Claims (5)

In the white light-based 3D scanner, the ambient light is detected using an illumination sensor, and the value is used to optimally adjust the brightness of the pattern irradiation device and the exposure time of the camera to suit the characteristics of the 3D restoration target object or environment. Way.
The method of claim 1,
The illuminance sensor is mounted outside the system to measure the illuminance of the external environment during initial system operation and periodically.
The measured illuminance value is compared with the threshold illuminance value set inside the system. The method of claim 2,
The critical illuminance inside the system is
It can be set externally and the set threshold illuminance is ranged between the maximum and minimum values,
If less than the minimum value, the projector's PWM Duty is reduced to reduce brightness and increase camera exposure,
How to increase the projector's PWM Duty to increase brightness and increase camera exposure if greater than maximum. The method of claim 2,
Illuminance detected by the illuminance sensor,
If you fall within the range of critical illuminance values inside the system,
Create a pattern for the projector,
A method of synchronizing the ON time of a projector light used to project a generated pattern onto an object with the exposure time of a camera. The method of claim 1,
Count the number of calculated points
Comparing with a set threshold error number;
When within the set threshold error number range,
Perform the final three-dimensional operations,
When out of the set threshold error number range,
Although the correspondence point increases constantly, if there is a lot of overlap, the camera exposure and the brightness control of the projector are changed again.
If the corresponding point does not increase constantly, it is determined by the shadow or discontinuous surface to perform the final three-dimensional operation step.

Images