CN102136143B - Focal length calibration method for single optical centre reflection and refraction camera based on spatial collinear point - Google Patents

Abstract

The invention discloses a focal length calibration method for a single optical centre reflected refraction camera based on the spatial collinear point, comprising the following steps of obtaining an image point set, randomly sampling, building and solving a constraint equation, determining the focal length and the like, wherein the focal length of the single optical centre reflected refraction camera is calibrated. In the focal length calibration method for a single optical centre reflected refraction camera based on the spatial collinear point, a polynomial constraint equation about the focal length is given by the reflected refraction projection of three collinear points of any space; the focal length is calibrated on the basis of the constraint equation without carrying out secondary curve fitting for linear images; and no information of a scene is required. The focal length calibration method has the advantages of high flexibility and high calibration precision; and under the noise level of three pixels, the average precision of 99.6% can be achieved.

Description

Focus Calibration Method of Single Optical Center Catadioptric Camera Based on Spatial Collinear Points

technical field

The invention relates to a method for calibrating the focal length of a camera, in particular to a method for calibrating the focal length of a single optical center catadioptric camera based on spatial collinear points.

Background technique

A large field of view imaging system is desirable for many computer vision applications, such as robot navigation, visual surveillance, virtual reality, etc. Ordinary cameras have a very limited field of view. The single optical center catadioptric camera is composed of an ordinary camera and a rotating mirror. The space scene is first projected onto the rotating mirror, and then reflected by the mirror and projected to the ordinary camera for imaging. Therefore, the single optical center catadioptric camera has a very large viewing angle. Camera calibration is to estimate the physical imaging parameters from three-dimensional scenes to two-dimensional images, which is the premise and basis of many vision applications. The focal length is the most important imaging parameter, and other imaging parameters can be obtained relatively easily from the prior information of the system configuration or the scene, and the focal length estimation is an essential content in camera calibration.

The calibration of a single optical center catadioptric camera can generally be divided into three categories: one is to use calibration objects for calibration, such as using one-dimensional calibration rods, calibration balls, three-dimensional calibration blocks, etc. This type of method has high calibration accuracy, but needs to make precise calibration objects, so the flexibility is poor and the scope of application is small; the second type does not require any scene information, and uses the point correspondence in multiple images to perform self-calibration. However, due to the large distortion of the catadioptric image, finding the point correspondence itself is a very difficult task, so the calibration accuracy is poor; the third category is based on the catadioptric projection of the straight line, and the straight line is an artificial scene. The ubiquitous geometric elements in , the catadioptric image of a single optical center of a straight line in space is a quadratic curve segment. This kind of method generally needs to fit the quadratic curve to the image of the straight line, and the calibration accuracy depends heavily on the curve fitting. accuracy. However, due to the occlusion of the specular surface, the image of a straight line in space only occupies a small section of the entire quadratic curve, which makes the curve fitting effect very poor, resulting in poor calibration accuracy.

The invention patent application with the application number CN201010222347.2 discloses a real-time method for calculating the focal length of a camera from an image sequence, which includes the following steps: selecting two adjacent images from the image sequence; characterizing the two images point matching to obtain matching point pairs; solving any seven groups of matching point pairs in the matching point pairs to obtain a plurality of possible solutions of the focal length; evaluating the accuracy of each possible solution and using the evaluation value as each The weight of the possible solution; calculate the weighted summation for each possible solution according to the weight to evaluate the probability density function of the focal distance; multiple sampling on the probability density function surface of the focal distance to obtain a plurality of sampling points; use A Gaussian function is used to fit the sampling points, and the peak value of the function is used as the final focal length value. The real-time method of calculating the camera focal length from the image sequence with the above steps has serious defects of poor flexibility and small scope of application.

Contents of the invention

The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method for calibrating the focal length of a single optical center catadioptric camera based on spatial collinear points with good flexibility, wide application range and high calibration accuracy.

In order to solve the problems in the prior art, the method for calibrating the focal length of a single optical center catadioptric camera based on spatially collinear points in the present invention includes the following steps:

1) Obtain image point set: extract the edge of the catadioptric image, and obtain an image point set belonging to a spatial straight line from the edge of the image;

2) Random sampling: randomly sample M groups of image points from the image point set obtained in step 1), each group of image points is composed of three different image points, wherein M≥1;

3) Establishment and solution of the constraint equation: determine the polynomial constraint equation of each group of image points sampled in step 2) about the focal length, and solve to obtain M focal length estimates;

4) Determination of the focal length: preprocess the M estimated focal lengths obtained in step 3), use the projective properties of the spatial points in the unit visual sphere imaging model, and determine and step respectively according to each focal length estimated value after preprocessing 1 The projection point set of the spatial point corresponding to the acquired image point set on the unit viewing sphere, using the projection point set to fit a plane passing through the center of the unit viewing sphere, and calculating the average distance from the projection point set to the plane, The focal length estimate corresponding to the minimum average distance is the determined focal length of the single optical center catadioptric camera.

In the above method, the edge extraction of the catadioptric image described in step 1) is through the edge extraction method based on Canny operator.

In the above method, the image point set belonging to a spatial straight line is obtained from the edge of the image in step 1) using a clustering method or a manual method.

In the above method, the sampled M groups of image points in step 2) are determined according to the criteria in the RANSAC method.

In the above method, the establishment of the constraint equation described in step 3) includes the following sub-steps:

则

是将图像原点移至图像主点p＝(u₀ v₀ 1)^T的平移变换，令{m_i：i＝1，2，3}表示步骤2)中采样的任意一组图像点，则

i＝1，2，3；3.1) Preprocessing: Let m=(u v 1) ^T represent any image point,

but

is the translation transformation that moves the origin of the image to the principal point of the image p=(u ₀ v ₀ 1) ^T , let {m _i :i=1, 2, 3} represent any set of image points sampled in step 2), but

i=1,2,3;

3.2) Establishment of polynomial constraint equation: the polynomial constraint equation of a group of image points {m _i : i=1, 2, 3} with respect to focal length f is expressed as follows:

$\left\{\begin{array}{c}{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&xi;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}\\ {\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}\sqrt{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}\sqrt{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}\sqrt{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="r\; f"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">r</figure-callout></span><figure-callout\; id="2"\; label="r\; f"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">\; </figure-callout>}{\mathrm{<span\; class="notranslate">\; </span>}}^{}{\mathrm{<span\; class="notranslate">f</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&xi;</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

in:

${\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="ix"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">ix</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; the\; s</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; +</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="iy"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">iy</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; the\; s</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}}$

${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; ix</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2,3</span>}$

${\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; det</span>}\left[\begin{array}{cc}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; x</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}\end{array}\right]<span\; class="notranslate">\; ,</span>$ ${\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; det</span>}\left[\begin{array}{cc}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; x</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}\end{array}\right]<span\; class="notranslate">\; ,</span>$ ${\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; det</span>}\left[\begin{array}{cc}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; x</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}\end{array}\right]$

${\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="ix"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">ix</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&tau;s</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; ix</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; the\; <figure-callout\; id="2"\; label="s"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">s</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

τ=(1-ξ ² )/ξ ²

r is the aspect ratio of the image, s is the tilt factor, and ξ is the mirror parameter;

When ξ=1, that is, when the mirror surface is a paraboloid of revolution, the polynomial constraint equation (1) is sorted into a 4th degree polynomial constraint equation about the focal length f;

When ξ≠1, that is, when the mirror surface is of other types, the polynomial constraint equation (1) is sorted into an 8th degree polynomial constraint equation about the focal length f.

In the above method, preprocessing the M focal length estimates obtained in step 3) as described in step 4) is to sort the M focal length estimates obtained in step 3), and discarding the N extreme focal length estimates before and after Value, where N<M/2.

The present invention is based on the single optical center catadioptric camera focal length calibration method based on space collinear points, and uses the catadioptric projection of any three collinear points in space to give a polynomial constraint equation about the focal length, and calibrates the focal length based on the constraint equation without The quadratic curve fitting of the straight line image does not require any information of the scene, it has greater flexibility, and the calibration accuracy is higher. Under the noise level of 3 pixels, the average accuracy can reach 99.6%.

Description of drawings

Fig. 1 is the overall flowchart of the focal length calibration method of the single optical center catadioptric camera based on the space collinear point of the present invention;

Fig. 2 is a schematic diagram of a unit viewing sphere imaging model of a single optical center catadioptric camera.

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings.

FIG. 1 is an overall flowchart of the method for calibrating the focal length of a single optical center catadioptric camera based on spatially collinear points in the present invention.

The method for calibrating the focal length of a single optical center catadioptric camera based on the spatial collinear points of the present invention will be described in detail below with reference to FIG. 1 .

Step S101, acquiring an image point set.

The catadioptric image of a single optical center of a straight line in space is a quadratic curve segment. Generally speaking, the image of a straight line is the edge of the image. Therefore, the edge extraction method based on the Canny operator is firstly used to extract the catadioptric image. The edge, and then use the clustering method or use some image processing software for manual punctuation to obtain the image point set belonging to a spatial straight line from the edge of the catadioptric image.

Step S102, random sampling.

M groups of image points are randomly sampled from the image point set obtained in step S101, each group of image points is composed of three different image points, where M≥1. The sampling of image points can be determined according to the criteria in the random sampling consistency RANSAC (RANdom SAmple Consensus) method, and there is a certain distance between a group of image points sampled each time.

Step S103, establishing and solving constraint equations.

Determine the polynomial constraint equation of each group of image points sampled in step S102 with respect to the focal length, and solve to obtain M focal length estimation values.

The establishment and solution process of the constraint equation in this step is introduced in detail as follows:

则

是将图像原点移至图像主点p＝(u₀ v₀ 1)^T的平移变换，令{m_i：i＝1，2，3}表示步骤S102中采样的任意一组图像点，则

i＝1，2，3。1) Preprocessing: Let m=(u v 1) ^T represent any image point,

but

is the translation transformation that moves the origin of the image to the principal point of the image p=(u ₀ v ₀ 1) ^T , let {m _i : i=1, 2, 3} represent any set of image points sampled in step S102, then

i=1,2,3.

2) Establishment of the constraint equation: the polynomial constraint equation of a group of image points {m _i : i=1, 2, 3} with respect to the focal length f is expressed as follows:

$\left\{\begin{array}{c}{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&xi;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}\\ {\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}\sqrt{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}\sqrt{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}\sqrt{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="r\; f"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">r</figure-callout></span><figure-callout\; id="2"\; label="r\; f"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">\; </figure-callout>}{\mathrm{<span\; class="notranslate">\; </span>}}^{}{\mathrm{<span\; class="notranslate">f</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&xi;</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

in:

${\mathrm{<span\; class="notranslate">\; \&eta;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="ix"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">ix</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; the\; s</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; +</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="iy"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">iy</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; the\; s</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}}$

${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; ix</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2,3</span>}$

${\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; det</span>}\left[\begin{array}{cc}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; x</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}\end{array}\right]<span\; class="notranslate">\; ,</span>$ ${\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; det</span>}\left[\begin{array}{cc}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; x</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}\end{array}\right]<span\; class="notranslate">\; ,</span>$ ${\mathrm{<span\; class="notranslate">\; D.</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; det</span>}\left[\begin{array}{cc}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; x</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; x</span>}}\\ {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}& {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; the\; y</span>}}\end{array}\right]$

${\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="ix"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">ix</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; f</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&tau;s</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; ix</span>}}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; iy</span>}}\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="iy"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">iy</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; the\; <figure-callout\; id="2"\; label="s"\; filenames="HDA0000050621890000021.png"\; state="\{\{state\}\}">s</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

τ=(1-ξ ² )/ξ ²

r is the aspect ratio of the image, s is the tilt factor, and ξ is the mirror parameter;

When ξ=1, that is, when the mirror surface is a paraboloid of revolution, the polynomial constraint equation (1) is sorted into a 4th degree polynomial constraint equation about the focal length f;

When ξ≠1, that is, when the mirror surface is of other types, the polynomial constraint equation (1) is sorted into an 8th degree polynomial constraint equation about the focal length f.

3) Polynomial constraint equation solution.

Using the symbolic calculation function of Maple to solve the analytical expression of the focal length f, there are 8 solutions to the polynomial equation of degree 8. Due to the existence and uniqueness of the focal length of the camera, only one of the 8 solutions is legal, and the others are imaginary numbers or close to zero. , therefore, for the M groups of image points collected in step S102, M focal length estimation values can be obtained.

Step S104, determining the focal length.

Before this step is introduced in detail, firstly a detailed introduction is given to the imaging model of the single optical center catadioptric camera involved in this step and the projective properties of the spatial points in the imaging model.

Fig. 2 is a schematic diagram of a unit viewing sphere imaging model of a single optical center catadioptric camera.

As shown in Figure 2, the imaging process of the unit viewing sphere imaging model of a single optical center catadioptric camera is as follows:

First, under the coordinate system O-xyz of the unit viewing sphere 1, with the center of the sphere O as the perspective center, the point X in the three-dimensional space is projected to the point X ^s on the unit viewing sphere 1;

Then, with the spatial point O ^c as the center, the point X ^s is projected by a virtual pinhole camera perspective onto the image plane 2 perpendicular to the straight line O ^c O to form an image point m. The point O ^c is called the virtual camera optical center, and the plane 2 Called the catadioptric image plane.

In the unit viewing sphere imaging model of the single optical center catadioptric camera shown in Fig. 2, the straight line O ^c O is the optical axis of the virtual camera, and the intersection point p=(u ₀ v ₀ 1) ^T between it and the image plane 2 is the image The principal point, the distance ξ=||OO ^C || from O ^c to O is called the mirror parameter. Different mirror parameters correspond to different types of mirror surfaces: when ξ=1, the mirror surface is a paraboloid of revolution; when 0<ξ<1, the mirror surface is a hyperboloid of revolution or an ellipsoid of revolution. Let K be the internal parameter matrix of the virtual pinhole camera, expressed as follows:

$\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; =</span>\left[\begin{array}{ccc}\mathrm{<span\; class="notranslate">\; rf</span>}& \mathrm{<span\; class="notranslate">\; the\; s</span>}& {\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; f</span>}& {\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right]$

where, f is the focal length, r is the aspect ratio, p=[u ₀ , v ₀ , 1] ^T is the principal point, s is the distortion factor, [f, r, s, u ₀ , v ₀ , ξ] is the single light Intrinsic parameters of the catadioptric camera. Any spatial straight line L is first projected onto a great circle on the unit viewing sphere 1, and the image of the great circle on the unit viewing sphere 1 under the virtual pinhole camera is a quadratic curve C. Therefore, the reflection and refraction of the spatial straight line L The image is a quadratic curve segment on the quadratic curve C, if and only if the space straight line L is coplanar with the virtual camera optical axis O ^c O, the quadratic curve segment degenerates into a straight line segment.

The projective properties of any spatial point in the unit viewing sphere imaging model of a single optical center catadioptric camera are as follows:

Suppose m=(u v 1) ^T is the catadioptric image point of space point X, under the virtual pinhole camera coordinate system O ^c -x ^c y ^c z ^c , the projection point X ^s of space point X on the unit viewing sphere 1 Expressed as follows:

${\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; \&xi;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\sqrt{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&tau;\&eta;</span>}}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}{\mathrm{<span\; class="notranslate">\; K</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; m</span>}$

If the mirror surface is a paraboloid of revolution, that is, ξ=1, then

${\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}{\mathrm{<span\; class="notranslate">\; K</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; m</span>}$

Wherein, τ=(1-ξ ² )/ξ ² , η=m ^T ωm, ω=K ^-T K ^-1 is the image of the absolute conic curve in the virtual pinhole camera.

On the basis of the detailed description of the unit viewing sphere imaging model based on the above-mentioned single optical center catadioptric camera and the projective properties of the spatial points in the imaging model, the determination of the focal length in this step is introduced in detail as follows:

1) Preprocessing, sorting the M estimated focal length values obtained in step S103, and discarding the N extreme focal length estimated values before and after, wherein N<M/2;

2) Utilizing the projective properties of the spatial points in the unit viewing sphere imaging model, according to each preprocessed focal length estimation value, respectively determine the projection point sets of the spatial points corresponding to the image point sets obtained in step S101 on the unit viewing sphere, Using the projected point set, fit a plane passing through the center of the unit viewing sphere, calculate the average distance from the projected point set to the plane, and the estimated value of the focal length corresponding to the minimum average distance is the determined single optical center catadioptric camera focal length.

In conclusion, the embodiments of the present invention disclose the preferred implementation thereof, but are not limited thereto. Those skilled in the art can easily comprehend the spirit of the present invention based on the above-mentioned embodiments, and make different extensions and changes, but as long as they do not deviate from the spirit of the present invention, they all fall within the protection scope of the present invention.

Claims (4)

1. the monochromatic light oculo cardiac reflex refraction focal length of camera scaling method based on the space collinear point is characterized in that, may further comprise the steps:

1) obtains the image point set: extract the reflected refraction edge of image, obtain belonging to the image point set of a space line from edge of image;

2) stochastic sampling: concentrate stochastic sampling M picture group picture point from the picture point that step 1) is obtained, every picture group picture point is made up of three pictures different points, wherein M >=1;

3) foundation of equation of constraint with find the solution: confirm step 2) in every picture group picture point of sampling about the polynomial expression equation of constraint of focal length, the foundation of said equation of constraint comprises as follows step by step:

3.1) pre-service: make m=(u v 1) ^TRepresent any picture point,

Then Be that image origin is moved to figure principal point p=(u ₀v ₀1) ^TTranslation transformation, make { m _i: i=1,2,3} representes step 2) middle any set of diagrams picture point of sampling, then

3.2) foundation of polynomial expression equation of constraint: set of diagrams picture point { m _i: i=1,2,3} representes as follows about the polynomial expression equation of constraint of focal distance f:

Wherein:

τ＝(1-ξ ²)/ξ ²

R is an aspect ratio, and s is an inclination factor, and ξ is the minute surface parameter;

When ξ=1, when promptly mirror surface was the paraboloid of revolution, polynomial expression equation of constraint (1) was organized into 4 order polynomial equation of constraint about focal distance f;

When ξ ≠ 1, when promptly mirror surface was other type, polynomial expression equation of constraint (1) then was organized into 8 order polynomial equation of constraint about focal distance f; And find the solution and obtain M focal length estimated value;

4) confirming of focal length: M focal length estimated value to obtaining in the step 3) carried out pre-service; M focal length estimated value to obtaining in the step 3) sorts; Each N extreme focal length estimated value before and after abandoning, N＜M/2 wherein, the projective properties of spatial point in the ball imaging model are looked by the unit of utilization; According to pretreated each focal length estimated value; The spatial point that definite respectively picture point set pair that obtains with step 1 is answered is looked the projection point set on the ball in unit, utilizes this projection point set, plane of looking the ball centre of sphere through unit of match; Calculate the mean distance of this projection point set to this plane, the focal length estimated value of corresponding minimum average B configuration distance is determined monochromatic light oculo cardiac reflex refraction focus of camera.

2. the monochromatic light oculo cardiac reflex refraction focal length of camera scaling method based on the space collinear point according to claim 1 is characterized in that extracting the reflected refraction edge of image described in the step 1) is through the edge extracting method based on the Canny operator.

3. the monochromatic light oculo cardiac reflex refraction focal length of camera scaling method based on the space collinear point according to claim 1; It is characterized in that the image point set that obtains belonging to a space line from the reflected refraction edge of image described in the step 1) is to use clustering method or manual mode.

4. the monochromatic light oculo cardiac reflex refraction focal length of camera scaling method based on the space collinear point according to claim 1 is characterized in that step 2) described in sampling M picture group picture point be to confirm according to the criterion in the RANSAC method.

Images