Laser triangulation is a 3D visual detection algorithm which is commonly used and easy to understand in the field of visual detection.

This paper mainly from the application level to elaborate, including the camera and laser type selection, the advantages and disadvantages of lap method analysis, software development in the process of matters needing attention.

I. Principle and

A single line of fine laser light is projected onto the surface of an object. Due to the change in the height of the surface of the object, the laser line is bent. According to the deformation of this line, the accurate three-dimensional contour of the surface of the object can be calculated.

As shown in the figure below, the basic structure includes: 1) 2D/3D camera 2) line laser 3) lens 4) fixing frame and mounting method

Ii. Characteristics

1) You can get both X and Z coordinates

2) There must be relative movement between the camera and the measured object

3) Mainly used for online 3D measurement

4) Suitable for short range, high precision and high speed measurement

3. Key parameters

3.1 Camera selection

You can choose a common array camera or a 3D camera to get 3D images or point cloud data. With a regular array camera, you need to extract the contours yourself and reconstruct the depth image by calibrating them. Halcon has a ready-made routine to do this. If the line frequency requirements are not high, Z-direction accuracy requirements are not high, high-speed array camera can be used to achieve.

My recent project has high requirements on line frequency and precision, so I still choose the 3D camera scheme. Germany's SICK and AT cameras are the two most popular 3D cameras used in industrial inspection applications, with the highest row frequency of dozens of KHZ.

The size of the video is not only related to the number of lines, but also directly related to the width of the ROI set (pixel points) and exposure time.

3.2 Selection of line laser

There are many evaluation parameters of line laser, such as uniformity, point stability, collimation, line of sight, power stability. The parameters of the laser itself are: fan Angle, power, depth of field and so on. In actual project selection, the most commonly used parameters are:

Fan Angle: The larger the fan Angle, the longer the laser line corresponding to the same working distance.

Power: The higher the power, the greater the intensity of the laser (the brighter the naked eye). For black non-reflective material, to choose a larger power of the laser. Power stability will also affect the sensitivity of measurement, poor power stability, will not be able to use a fixed threshold method, for low contrast object measurement becomes difficult

Uniformity: Poor uniformity can reduce resolution and accuracy

After a number of different brands of laser testing, Germany's ZLaser laser laser is a cost-effective laser, a variety of models can meet the application of different occasions.

3.3 Selection of lap mode

1) Standard installation: the laser is vertical to the material plane, and the camera is at an α Angle to the laser

Suitable for most situations

Advantages: The points on the outline have the same Y coordinate, simple calibration

Disadvantages: There are blind spots

2) Reverse installation: the camera is vertical to the material plane, and the laser and the camera are at an Angle of α

Application: Flat objects

Advantages: Increased high resolution

Disadvantages: The Y coordinates of the points on the outline are different, and the calibration is complicated

3) Launch installation or open field installation

Application: It is suitable for flat objects with weak return light. Due to direct reflection, can increase the brightness of the object back light (for some materials, may be a disadvantage).

Advantages: Greatly increased high resolution

Disadvantages: Calibration is complicated

4) Dark field installation

Application: plane object with strong return light.

Advantage: Can reduce direct light reflection

Disadvantages: can reduce the high resolution, calibration is complicated

3.4 Selection of measurement Angle

Larger measurement Angle can bring higher Z-direction resolution, but also lead to larger blind area. So you need to make trade-offs based on the actual project situation.

For example: 5mm high object, blind area is 4.2mm

Iv. Practical application

4.1 Application Examples

The following information needs to be known:

Object size (length, width, height)

X, Y, Z direction accuracy requirements

Speed of scanning

Application Type

The material

For example, if the size of the object to be measured is 80*50*5 (length * width * height), the x-direction accuracy is 0.3mm, Y-direction accuracy is 0.3mm, and the scanning speed is 2m/s, then the X-direction resolution should not be higher than 0.1mm/pixel, and the Y-direction resolution should not be higher than 0.1mm/pixel.

Camera Selection

The number of pixels to be selected is not less than 80/0.1=800; The corresponding line frequency is not less than 2000/0.1=20KHZ.

The maximum line frequency the camera can achieve depends on the size of ROI effective region, exposure time and contour extraction algorithm. The effective area of ROI depends on the thickness range of the material, and the exposure time partly depends on the material (some materials, such as rubber tires, are black absorbent materials, and the exposure time is higher when the brightness of the light source is the same).

2) Lens selection

This is the same thing as 2D. It mainly depends on the working distance of the field installation. The larger the focal length, the longer the working distance

3) Lap method

As mentioned above, each has its own advantages and disadvantages, which method can be selected according to the difficulty of calibration, detection accuracy, material of flat objects and other factors

4.2 Drawing

When the camera and laser Angle are fixed, the difference in the orientation of the camera can lead to the difference in gray polarity (that is, the higher the height of the object, the brighter or darker the corresponding gray value). Simple summary: The camera itself is installed in a direction, if the laser in the positive direction of the camera above, the higher the height of the object, the lower the corresponding gray value; Otherwise, the corresponding gray value is higher.

Based on this, if you want to change the grayscale polarity, you can do so by setting the ReverseY parameter without changing the camera mounting direction.

4.3 Calibration

The purpose of calibration: to obtain the inside and outside parameters of the camera and the direction of the laser light plane. We can then calculate the physical unit size of the object in the X and Z directions. 3D cameras are generally integrated with a ready-made calibration module, so it is relatively easy to calibrate. The calibration method often used is: zigzag calibration plate

By moving the calibration block in the motion direction and calling the calibration algorithm of the camera itself, the 3D camera calibration is realized.

Summary: Laser triangulation is a very mature algorithm in the field of visual detection. It can be used in various fields to achieve visual 3d detection. Huayan Tianchuang is a leader in the field of visual detection.