How to calibrate the camera

How to monitor the deformation of buildings by photogrammetry

"Zhu Xun China" network answers for you:

In recent years, the close-range photogrammetry method has been applied to measure the settlement of foundation and building, the inclination of building, the parameters of cracks, the parameters of model deformation state, and the deformation of construction machinery parts. Close-range photogrammetry usually uses phototheodolite, ordinary camera or high-speed camera to continuously record a large number of point deformation information of buildings and test models in a certain period of time or in an instant through vertical, equidistant and intersecting photography. And use three-dimensional coordinate measuring instrument, electronic computer, precision three-dimensional surveying and mapping instrument or analytical surveying and mapping instrument to measure the two-dimensional or three-dimensional relative deformation of observation points with time according to analytical method or analog analytical method. Photographs taken can also be checked and measured at any time as files.

What is the difference between aerial digital imaging system and ordinary digital camera in calibration?

Aerial photogrammetry (also known as aerial digital imaging system) is an ancient subject, which uses a special camera installed on the plane to shoot the ground at a certain height. Of course, there are many technical requirements. Using photographic data, technicians can carry out corresponding topographic survey and other technical analysis.

The principle of aerial digital imaging system is the same as that of ordinary digital camera in calibration, but the shooting range of aerial digital imaging system is very wide. In addition to the calibration of ordinary digital cameras, we should also consider the earth's surface correction, because the earth's surface is not flat. Please refer to the books on measurement for details.

In order to realize the inspection of ordinary digital cameras before photogrammetry, a simple method and process of establishing indoor digital camera calibration field are proposed. Firstly, the calibration site is established, including the selection of site, the production of landmark points and the measurement of their coordinates, and then the reliability of the calibration site is verified. The results show that the simple indoor digital camera calibration field can meet the calibration requirements of ordinary digital cameras in close-range photogrammetry. Moreover, the calibration site is simple to build and the cost is much lower than that of the traditional digital camera calibration site, which is an improvement on the traditional large-scale indoor digital camera calibration site.

What role does the directional element play in the coordinate system transformation of general photogrammetry?

Directional elements play a role in determining the position in the coordinate system transformation of general photogrammetry.

The internal orientation element is a parameter that determines the relationship between the camera lens center (strictly speaking, it should be the image node of the lens) and the image position. It includes three parameters: the positions x and y of the main point of the image (the vertical foot of the main optical axis on the image plane) relative to the image center, and the vertical distance f (the main distance) from the lens center to the image plane, which is generally determined by camera calibration.

Determination of internal orientation elements

This is done by camera calibration. Because of the diversity of cameras and calibration methods, camera calibration is not standardized. According to different calibration environments, the commonly used calibration methods include control field-based calibration method (optical laboratory calibration method, experimental field calibration method), in-service calibration method (that is, camera calibration while completing a certain measurement task), self-checking calibration method, star point calibration method and vanishing point-based calibration method.

How to test the pixels of a camera

Resolution wedge diagram

1. Resolution is an important parameter to measure the quality of a camera. It refers to the maximum number of lines that people can see on the monitor when the camera shoots black and white stripes arranged at equal intervals. When it exceeds this number, only a piece of gray can be seen on the screen, and black and white lines can no longer be distinguished. Clarity is divided into horizontal resolution and vertical resolution.

2. Test method: The camera takes a comprehensive test chart, and visually observes the maximum number of distinguishable lines on the central wedge of the image or the maximum number of distinguishable lines of ten groups of central definition line segments on the monitor.

3. Matters needing attention during the examination

(1) Use a lens with good imaging quality, because the quality of the lens will affect the final test result.

(2) When displaying, use a black-and-white display with more than 600 lines. If a color monitor is used, adjust the color saturation knob to the lowest level to avoid the interference of chromaticity signal on brightness signal.

4, should be paid attention to when purchasing

(1) Use Sony and Panasonic original cameras for horizontal comparison, and observe the differences between the two cameras in distinguishing black and white line groups; The performance index of the original installation machine is true and reliable. By comparison, we can draw a correct conclusion on the definition index of the camera to be purchased.

(2) When buying a single-board computer, sometimes the matching lens has poor imaging quality. In addition to the resolution of the test center, the resolution of four corners should be tested, and there should be no ambiguity and deformation. Otherwise, a better lens should be replaced.

Second, the minimum illumination index should meet the relevant conditions.

1, the concept of minimum illumination, the minimum illumination of the subject when the brightness output level generated by the camera is half of the rated level (700mv).

2. Detection method

Ten-level gray scale test card

(1) comparison method: put the camera in a dark room and choose an original camera from a famous factory.

In contrast, two manual aperture lenses of the same model are used, and the darkroom is equipped with a 220V incandescent lamp controlled by a voltage regulator. The voltage can be adjusted from 0V to 220V, and the indoor illumination can also be adjusted from the darkest to the brightest. The two cameras are aimed at objects with rich layers, and the indoor brightness is reduced until the darkness of the object cannot be seen clearly, or the aperture of the first lens is reduced for comparison. According to the original camera standard of a famous factory,

(2) Instrumental method: darkroom test: aim the camera at the 10 gray scale test card, reduce the indoor brightness until the video signal output by the camera drops to 350mv on the oscilloscope, and then measure the illuminance value on the surface of the test card with an exposure meter to calculate the minimum illuminance.

3. It should be noted that the value of minimum illumination is related to the following four factors.

(1) lens aperture

(2) Color temperature of light source

(3) the amplitude of the video signal

(4) reflectivity (reflectivity and background of the target)

Only when the above four related conditions are marked, the minimum illumination of the test is meaningful. We can't simply compare the nominal illuminance values of one brand camera and another brand camera without the above four test conditions, otherwise we can't draw a conclusion that which camera has better low illuminance characteristics.

4, should be paid attention to when purchasing

Choose illumination camera: the nominal minimum illumination can only be used as a reference, and the key is to choose a suitable camera according to the needs of the use. Use an exposure meter to measure the minimum illumination at the scene. Simulate the lighting environment in a dark room, choose different cameras to test or use a large aperture lens until the ten-level outline of the gray test card is clear and recognizable.

Third, the relationship between signal-to-noise ratio and illumination

1, the so-called "signal-to-noise ratio" refers to the ratio of signal voltage to noise voltage, which is usually represented by symbol S/N. The signal-to-noise ratio can be divided into brightness signal-to-noise ratio and chroma signal-to-noise ratio. Signal-to-noise ratio is also a main parameter of the camera.

When the camera shoots a bright scene, the picture displayed by the monitor is usually bright, and it is difficult for observers to see the interference noise in the picture; When the camera shoots a dark scene, the picture displayed by the monitor is dim, and the observer can easily see the snowflake-like interference noise in the picture at this time. The higher the signal-to-noise ratio of the camera, the smaller the influence of interference noise on the picture.

2. Detection method of brightness signal-to-noise ratio

(1) simple discriminant method

Signal-to-noise ratio of brightness: close the lens aperture or cover the lens to observe the amount of snowflake interference noise on the display.

Chromaticity SNR: aim the camera at the white balance test card and observe the number of colored noise points.

(2) Contrast method

Put the camera in a dark room, select the original camera of a well-known manufacturer for comparison, use two manual aperture lenses of the same model, aim the camera to be tested and the original camera at the black balance test card, adjust the illumination with a dimmer until the snow-like interference noise appears obviously on the picture, compare the density and size of noise points, and estimate the signal-to-noise ratio.

(3) Instrumental methods

Brightness SNR: aim the camera at the 10 gray scale test card, adjust the aperture to make the video level output by the camera reach 350mv, connect the signal to the video noise meter, and directly read the SNR reading on the dashboard.

3. Matters needing attention during the examination

The selection of (1) filter generally adopts 100KHZ low-pass and fg Qualcomm, and does not adopt video weighting curve.

② The influence of camera AGC and R correction should be considered.

4, should be paid attention to when purchasing

Due to the high price of video noise measuring instrument, domestic manufacturers generally do not have such equipment. Large manufacturers estimate the signal-to-noise ratio with oscilloscopes, while small manufacturers visually measure the Xue Huadian of the camera and fill in a number casually, so it is impossible to measure the chromaticity noise.

Therefore, when choosing a camera, we should choose the camera according to the lighting conditions of the site, measure and record it with an exposure meter, and use different illumination values of the site during the monitoring period. Go back to the office darkroom to simulate the different illumination values at the scene, observe which camera has more noise, increase the aperture of the lens or use a lens with large aperture and large luminous flux, or increase the brightness of the lights at the scene until the image quality displayed on the display reaches an acceptable level. This is your suitable camera.