Binocular vision synchronous positioning method and system

Abstract

The present invention discloses a binocular vision synchronous positioning method and system, which relates to the field of binocular vision positioning systems. The binocular vision synchronous positioning method includes providing a binocular camera and a host computer for controlling the working state of the binocular camera, wherein the binocular camera includes a left camera and a right camera, and the host computer synchronously calibrates the left camera and the right camera to obtain a delay frame number δ representing the delay time when the left camera image and the right camera image are transmitted to the host computer after the synchronous calibration. During visual positioning, the host computer captures the left camera image and the right camera image of the current frame, selects the left camera image or the right camera image of the current frame as a reference image based on the delay frame number δ, and searches for a synchronous frame image collected at the same time as the reference image. The host computer performs visual positioning based on the reference image and the synchronous frame image and outputs the target positioning result, which effectively improves the positioning accuracy of binocular vision.

Description

Technical Field

[0001] The invention relates to the field of binocular vision positioning systems, and in particular to a binocular vision synchronous positioning method and system. Background Art

[0002] With the development of computer technology, machine vision is increasingly used in the fields of industry, security, etc. Machine vision technology can automatically identify and measure targets, providing perceptual input for subsequent task processing. At present, the image input for visual recognition and positioning mainly comes from cameras. Generally, a single camera can be used to identify and locate some known specific cooperative targets, while binocular vision is generally used for identification and positioning of non-cooperative targets.

[0003] Binocular vision positioning systems generally include two cameras arranged side by side. The principle of binocular vision positioning is to observe the same object through two cameras at the same time. Since the positions of the two cameras are different, the images collected by the two cameras are also different. The difference between the two camera images is called parallax, which is the basis of binocular vision positioning. When the computer receives different images from the two cameras, it analyzes, matches and calculates these image features to determine the position and distance of the target object.

[0004] However, many binocular vision positioning systems directly use two network cameras for integration, and the back-end computer directly obtains the video data of the current two cameras for identification and positioning. Although this binocular vision positioning system is simple to develop, the video images output by the two cameras received by the back-end processor will not be synchronized, which will lead to inaccurate subsequent identification and positioning. In addition, there is a certain lag between the positioning result output by the binocular vision positioning system and the true value. The main reason for the lag is that camera imaging, image encoding and decoding, network transmission, image processing and other links all require a certain amount of time, and the current visual positioning system generally cannot provide the lag time to the outside world. Therefore, in order to improve the positioning accuracy of the binocular vision positioning system and provide prior parameters for subsequent applications, the binocular cameras need to be synchronized and the lag time needs to be provided. Summary of the invention

[0005] In view of the above problems and technical requirements, the inventors have proposed a binocular vision synchronous positioning method and system. The technical solution of the present invention is as follows:

[0006] A binocular vision synchronous positioning method includes providing a binocular camera and a host computer for controlling the working state of the binocular camera, wherein the binocular camera includes a left camera and a right camera;

[0007] Based on the left camera image captured by the left camera and the right camera image captured by the right camera, the host computer synchronously calibrates the left camera and the right camera, so as to obtain the delay frame number δ representing the delay time when the left camera image and the right camera image are transmitted to the host computer after the synchronous calibration;

[0008] During visual positioning, the host computer captures the left camera image and the right camera image of the current frame, selects the left camera image or the right camera image of the current frame as the reference image based on the delay frame number δ, and searches for the synchronous frame image acquired at the same time as the reference image. The host computer performs visual positioning based on the reference image and the synchronous frame image and outputs the target positioning result.

[0009] A further technical solution is to synchronously calibrate the left camera and the right camera, including:

[0010] The host computer controls the synchronous calibration device to generate calibration features so that the calibration features fall into the field of view of the left camera and the right camera at the same time;

[0011] When it is determined that the calibration features exist in both the left camera image and the right camera image, the host computer searches the acquired left camera image and the right camera image to obtain the corresponding frame sequence m and frame sequence n when the calibration features first appear in the left camera image and the right camera image;

[0012] Based on the frame sequence m and the frame sequence n, a delay frame number δ is generated in the host computer.

[0013] A further technical solution is that when performing binocular vision synchronous positioning, the host computer also outputs the lag time t d , in which, while controlling the synchronous calibration device to generate the calibration feature, the host computer records the initial time t of the lag calculation o ;

[0014] While outputting the target positioning result, the host computer records the lag calculation completion time t c ;

[0015] Calculate the initial time t based on the hysteresis o and hysteresis calculation completion time t c , host computer output lag time t d .

[0016] Its further technical solution is that when using binocular vision for synchronous positioning, the host computer is based on the interval period T s The left camera and the right camera are calibrated synchronously, where:

[0017] In each interval period Ts, after synchronous calibration, the delay frame number δ of the current interval period Ts is obtained in the host computer.

[0018] Its further technical solution is that, within the host computer, based on the lag time t d Adjust the interval period T s , then there is:

[0019]

[0020] Wherein, T max is the upper limit of the interval period, T min is the lower limit of the interval period, t d [p] is the lag time output by the host computer for the pth time, η is the fluctuation coefficient of the lag time t d and N is the sample coefficient of the lag time t d and t d is the sample average value of the lag time t d .

[0021] Its further technical solution is that the interval period T s is an integer, and the upper limit T max of the interval period and the lower limit T min of the interval period are also integers, wherein, 100 < Tmin < 200 and 300 < T max < 400.

[0022] Its further technical solution is that during initialization, the host computer sets the initial values of the upper limit T max of the interval period, the lower limit T min of the interval period, the interval period Ts and the number of delayed frames δ, wherein, Ts = T min and δ = 0.

[0023] Its further technical solution is that the binocular vision synchronous positioning system includes a synchronous calibration device for generating calibration features, a binocular camera for collecting image information, and a host computer for controlling the corresponding working states of the synchronous calibration device and the binocular camera. Among them, the host computer cooperates with the binocular camera and the synchronous calibration device to implement the above-mentioned binocular vision synchronous positioning method.

[0024] Its further technical solution is that the synchronous calibration device includes a laser. The laser is installed between the left-eye camera and the right-eye camera and generates laser light as calibration features, and the laser light simultaneously falls within the field of view of the left-eye camera and the right-eye camera.

[0025] Its further technical solution is that the binocular vision synchronous positioning system further includes a mechanical bracket, and the laser, the left-eye camera and the right-eye camera are fixed on the mechanical bracket.

[0026] The beneficial technical effects of the present invention are:

[0027] The synchronous calibration device is used to generate calibration features, and the left and right cameras are synchronously calibrated to obtain the delay frame number δ. Based on the delay frame number δ, the left camera image or the right camera image of the current frame is selected as the reference image, and the synchronous frame image collected at the same time as the reference image is searched. The host computer performs visual positioning based on the reference image and the synchronous frame image and outputs the target positioning result, which effectively improves the positioning accuracy. The lag time t of the binocular vision synchronous positioning is output while the binocular camera image is synchronously calibrated d , providing prior parameters for subsequent applications.

[0028] Based on the interval period T s When performing visual positioning, the interval period Ts is adjusted in an adaptive manner based on the lag time t d Adjust so that the lag time t d The synchronous calibration frequency is reduced under stable conditions, which reduces the interference of calibration features on the imaging of the binocular vision synchronous positioning system, reduces the amount of calculation, and improves the efficiency of visual positioning; at the lag time t d The calibration frequency under unstable conditions is increased, which improves the timeliness of the synchronous calibration of the binocular vision synchronous positioning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a structural schematic diagram of an embodiment of the binocular vision synchronous positioning system provided by the present invention.

[0030] Figure 2 It is a program flow chart of an embodiment of the binocular vision synchronous positioning method provided by the present invention.

[0031] Description of reference numerals: 101 - left camera, 102 - right camera, 103 - laser, 104 - mechanical support, 105 - host computer, 106 - communication link, 107 - control cable. DETAILED DESCRIPTION

[0032] The specific implementation of the present invention will be further described below in conjunction with the accompanying drawings.

[0033] In order to effectively improve the positioning accuracy of binocular vision positioning, the present invention proposes a binocular vision synchronous positioning method, including providing a binocular camera and a host computer 105 for controlling the working state of the binocular camera, wherein the binocular camera includes a left camera 101 and a right camera 102;

[0034] Based on the left camera image captured by the left camera 101 and the right camera image captured by the right camera 102, the host computer 105 synchronously calibrates the left camera 101 and the right camera 102, so as to obtain the delay frame number δ representing the delay time when the left camera image and the right camera image are transmitted to the host computer 105 after the synchronous calibration;

[0035] During visual positioning, the host computer 105 captures the left camera image and the right camera image of the current frame, selects the left camera image or the right camera image of the current frame as the reference image based on the delay frame number δ, and searches for the synchronous frame image collected at the same time as the reference image. The host computer 105 performs visual positioning based on the reference image and the synchronous frame image and outputs the target positioning result.

[0036] Specifically, a binocular camera generally includes a left camera 101 and a right camera 102 arranged side by side. The left camera 101 and the right camera 102 simultaneously capture images of the same target object. The left camera 101 captures and generates a left camera image, and the right camera 102 captures and generates a right camera image. The host computer 105 can generally be a computer, and the host computer controls the left camera 101 and the right camera 102 to simultaneously capture and transmit images.

[0037] In theory, the binocular vision system performs visual positioning through the left camera image and the right camera image collected at the same time, but in actual application, the transmission time required for the left camera 101 and the right camera 102 to transmit the image to the host computer 105 is not exactly the same, and the left camera image and the right camera image simultaneously received by the host computer 105 are not collected at the same time, resulting in inaccurate binocular vision positioning. Among them, the difference in the transmission time required for the left camera 101 and the right camera 102 to upload the image to the host computer is the delay time, which can be represented by the delay frame number δ between the left camera image and the right camera image simultaneously received by the host computer 105.

[0038] In order to determine the delay frame number δ representing the delay time, the left camera 101 and the right camera 102 need to be synchronously calibrated to obtain the delay frame number δ after calibration. The host computer 105 receives the left camera image and the right camera image, and sorts them by frame respectively, forming a left camera image set and a right camera image set accordingly. During visual positioning, the host computer 105 captures the left camera image and the right camera image of the current frame in real time, and based on the process of obtaining the delay frame number δ, selects one of the left camera image or the right camera image of the current frame as the reference image, and after selecting the reference image, searches for the synchronous frame image collected at the same time as the reference image in the corresponding left camera image set or the right camera image set.

[0039] Among them, the reference image and the synchronization frame image belong to different image sets. For example, when the current frame left-eye camera image is selected as the reference image, the upper computer 105 searches for the synchronization frame image in the right-eye camera image set. Conversely, when the current frame right-eye camera image is selected as the reference image, the upper computer 105 searches for the synchronization frame image in the left-eye camera image set. The specific method for determining the reference image and the synchronization frame image can be referred to the following implementation.

[0040] Since the reference image and the synchronous frame image can be considered as two images collected at the same time, visual positioning is performed based on the reference image and the synchronous frame image and the target positioning result is output, which effectively improves the positioning accuracy of binocular visual positioning. Among them, the method of visual positioning based on the reference image and the synchronous frame image is consistent with the binocular visual positioning method commonly used in the technical field.

[0041] Furthermore, when the left-eye camera 101 and the right-eye camera 102 are synchronously calibrated, the host computer 105 controls the synchronous calibration device to generate calibration features, so that the calibration features fall into the field of view of the left-eye camera 101 and the right-eye camera 102 at the same time;

[0042] When it is determined that the calibration features exist in both the left camera image and the right camera image, the host computer 105 searches the collected left camera image and the right camera image to obtain the corresponding frame sequence m and frame sequence n when the calibration features first appear in the left camera image and the right camera image;

[0043] Based on the frame sequence m and the frame sequence n, a delay frame number δ is generated in the host computer.

[0044] Specifically, when determining the delay frame number δ, the host computer 105 needs to search for the left camera image and the right camera image collected at the same time in the received image. Generally, the host computer 105 needs to search for the left camera image and the right camera image collected at the same time based on the appearance of the calibration feature. In order to provide the calibration feature, the host computer 105 controls the synchronous calibration device to generate the calibration feature, and makes the calibration feature fall into the field of view of the left camera 101 and the right camera 102 at the same time.

[0045] After the calibration features generated by the synchronous calibration device fall into the field of view of the left camera 101 and the right camera 102 at the same time, the calibration features will exist in the captured left camera image and the right camera image. Since it takes time for the left camera 101 and the right camera 102 to transmit images to the host computer 105, the host computer 105 also needs to determine whether the received left camera image and right camera image both have calibration features, so as to ensure that the left camera image and the right camera image captured at the same time can be searched based on the appearance of the calibration features. The host computer 105 can use existing commonly used technical means to determine whether the left camera image and the right camera image both have calibration features.

[0046] The host computer 105 obtains the frame sequence m corresponding to the first appearance of the calibration feature in the left camera image in the left camera image set; obtains the frame sequence n corresponding to the first appearance of the calibration feature in the right camera image in the right camera image set, wherein the delay frame number δ is the difference between the frame sequence m and the frame sequence n, which can be expressed as δ=|mn|.

[0047] Specifically, when determining the reference image and the synchronization frame image, the host computer 105 first captures the left camera image and the right camera image of the current frame. If m≥n, the left camera image of the current frame is selected as the reference image, and the current frame is the latest frame of the left camera 101 and the right camera 102, that is, the i-th frame. The i-δth frame is searched in the captured right camera image as the synchronization frame image, and visual positioning is performed based on the i-th frame of the left camera image and the i-δth right camera image, and the target positioning result is output;

[0048] If m<n, the right camera image of the current i-th frame is selected as the reference image, and the i-δth frame is searched in the acquired left camera image as the synchronization frame image. Visual positioning is performed based on the right camera image of the i-th frame and the i-δth left camera image, and the target positioning result is output.

[0049] Furthermore, when performing binocular vision synchronous positioning, the host computer also outputs the lag time t d , wherein, while controlling the synchronous calibration device to generate the calibration feature, the host computer 105 records the initial time t of the hysteresis calculation o ;

[0050] While outputting the target positioning result, the host computer 105 records the hysteresis calculation completion time t c ;

[0051] The initial time t is calculated based on the hysteresis o and the hysteresis calculation is completed at time t c , host computer 105 output lag time t d , the lag time t d Compared with the lag time t described in the background technologyd The lag time can be expressed as t d =t c -t o .

[0052] When using binocular vision synchronous positioning, the left camera 101 and the right camera 102 can be synchronously calibrated continuously or intermittently. Preferably, in one embodiment of the present invention, based on the interval period T s The left camera 101 and the right camera 102 are synchronously calibrated intermittently, which not only reduces the interference of the calibration features on the imaging of the left camera 101 and the right camera 102, but also reduces the amount of calculation and improves the efficiency of visual positioning. In each interval period Ts, after synchronous calibration, the current interval period T is obtained in the host computer 105. s The number of delayed frames δ. That is, each interval period T s Update the delay frame number δ once every 1 second, and in this interval period T S The reference image and the synchronous frame image are obtained based on the delayed frame number δ for visual positioning, and the target positioning result is output. And in each interval period T s When the delayed frame number δ is obtained within the interval period T, the left camera image and the right camera image are searched. s The corresponding frame sequence where the calibration feature first appears within .

[0053] Based on the interval period T s When the left camera 101 and the right camera 102 are synchronously calibrated intermittently, the interval period T s It can be a fixed value or a variable value. Preferably, in an embodiment of the present invention, a variable interval period Ts is used to synchronously calibrate the left camera 101 and the right camera 102 intermittently.

[0054] Furthermore, the variable interval period T s Adopting adaptive adjustment method, based on each interval period T s Internal output delay time t d Adjustment is made, then:

[0055]

[0056] Among them, T max is the upper limit of the interval period, T min is the lower limit of the interval period, t d [p] is the lag time of the pth output of the host computer, η is the lag time t d The fluctuation coefficient, N is the lag time t d The sample coefficient, t d is the lag time t d The sample mean of .

[0057] Specifically, the interval period T s , Interval cycle upper limit T max and the lower limit of the interval period T min It is expressed as an integer multiple of the period. The upper limit of the interval period T max and the lower limit of the interval period T min Set it when the host computer is initialized, and 100 <T min <200,300 <T max <400, in specific implementation, the upper limit of the interval period T max and the lower limit of the interval period T min The selection can be made according to the actual needs of the interval period Ts. In one embodiment of the present invention, the sample coefficient N is 20. Optionally, N can also be other positive integer values ​​to meet the requirements of the interval period Ts. S The adaptive adjustment shall prevail.

[0058] Due to the lag time t d Including the time for image transmission and processing, the lag time t d The fluctuation of also reflects the fluctuation of image transmission and processing time. According to the above formula based on the lag time t d For the interval period T s By making adjustments, when the image transmission and processing time are stable, the synchronous calibration frequency can be reduced, the interference of calibration features on imaging can be reduced, the amount of calculation can be reduced, and the efficiency of visual positioning can be improved; when the image transmission and processing time are unstable, the synchronous calibration frequency can be increased to improve the timeliness of synchronous calibration.

[0059] In order to realize the above-mentioned binocular vision synchronous positioning method, the present invention proposes a binocular vision synchronous positioning system, which includes a synchronous calibration device for generating calibration features, a binocular camera for collecting image information, and a host computer for controlling the corresponding working states of the synchronous calibration device and the binocular camera, wherein the host computer cooperates with the binocular camera and the synchronous calibration device to implement the above-mentioned binocular vision synchronous positioning method.

[0060] like Figure 1 As shown, the binocular camera includes a left camera 101 and a right camera 102 arranged side by side, and the left camera 101 and the right camera 102 cooperate with a synchronous calibration device so that the left camera 101 and the right camera 102 can simultaneously collect calibration features generated by the synchronous calibration device.

[0061] In one embodiment of the present invention, the synchronous calibration device is a laser 103, Figure 11 shows an installation method of the laser 103, wherein the laser 103 is installed between the left eye camera 101 and the right eye camera 102, and is arranged side by side with the left eye camera 101 and the right eye camera 102, and generates laser light as a calibration feature, and the laser light falls into the field of view of the left eye camera 101 and the right eye camera 102 at the same time. The host computer 105 is connected to the left eye camera 101 and the right eye camera 102 through a communication link 106, and obtains and processes the images of the left eye camera 101 and the right eye camera 102. The host computer 105 is connected to the laser 103 through a control cable 107 to control the opening and closing of the laser 103.

[0062] The binocular vision synchronous positioning system further includes a mechanical support 104 , and the laser 103 , the left-eye camera 101 , and the right-eye camera 102 are fixed on the mechanical support 104 .

[0063] In one embodiment of the present invention, using Figure 1 The binocular vision synchronous positioning system shown implements the binocular vision synchronous positioning method in the upper computer 105 in the form of a software program, wherein when the upper computer 105 forms an image set, the frame sequence i of the left camera image and the right camera image is accumulated in sequence from i=1 until the binocular vision synchronous positioning system stops working. Figure 2 A flowchart of a program of an embodiment of the present invention is shown, comprising the following steps:

[0064] Step 1: The host computer 105 initializes the program parameters and sets the upper limit of the interval period T max and the lower limit of the interval period T min , let the delay frame number δ=0, T s =T min , i=1, p=1, num=0, m=n=0, where num is the program loop count and the interval period upper limit T max and the lower limit of the interval period T min The value range of is the same as above.

[0065] Step 2: The host computer 105 captures the current frame images of the left camera 101 and the right camera 102, and marks the two frames as left camera images L i and the right camera image R i .

[0066] Step 3: The program loop count num increases by 1;

[0067] The host computer 105 determines the program cycle count and the interval period T S If num≥T SIf the laser 103 is in the off state, the host computer 105 controls the laser 103 to turn on and records the system time of the host computer 105 at this time as the initial time t for lag calculation. o The host computer 105 jumps to execute step 10.

[0068] If num≥Ts and the laser 103 is in the on state, the host computer 105 continues to execute step 4.

[0069] If num<Ts, the host computer 105 jumps to execute step 10.

[0070] Step 4: The host computer 105 processes the left-eye camera image L i and the right-eye camera image R i , and determines whether there are laser light features in both images. If there are laser light features in both the left-eye camera image L i and the right-eye camera image R i , the host computer 105 searches for the collected left-eye camera image and right-eye camera image, obtains the corresponding frame numbers m and n when the calibration features first appear in the left-eye camera image and the right-eye camera image, and then the host computer 105 executes step 5; otherwise, the host computer 105 jumps to execute step 10.

[0071] Step 5: The host computer 105 sets num = 0, calculates the delay frame number δ, and turns off the laser 103 to avoid long-term interference of the laser light on visual positioning.

[0072] Step 6: The host computer 105 determines the size relationship between the frame number m and the frame number n. If m≥n, the host computer 105 obtains the right-eye camera image of the (i - δ)th frame and marks it as R i-δ , and based on the left-eye camera image L i and the right-eye camera image R i-δ performs binocular vision synchronous positioning, and records the system time of the host computer 105 when the calculation is completed as the lag calculation completion time t c ;

[0073] If m < n, the host computer 105 obtains the left-eye camera image of the (i - δ)th frame and marks it as L i-δ , and based on the left-eye camera image L i-δ and the right-eye camera image R i performs binocular vision synchronous positioning, and records the system time of the host computer 105 when the calculation is completed as the lag calculation completion time t c .

[0074] Step 7: The host computer 105 calculates and outputs the lag time t of the binocular vision synchronous positioning system d , the pth lag time is t d [p], the lag time t d[p] is calculated in the same way as above.

[0075] Step 8: The host computer 105 calculates and updates the interval period T s , interval period T s The calculation method of is the same as above.

[0076] Step 9: p is incremented by 1, and the host computer 105 jumps to execute step 11.

[0077] Step 10: The host computer 105 determines the size relationship between the frame sequence m and the frame sequence n. If m ≥ n, the host computer 105 obtains the right camera image of the i-δth frame and marks it as R i-δ , based on the left camera image L i and the right camera image R i-δ Perform binocular vision synchronous positioning;

[0078] If m<n, the host computer 105 obtains the left camera image of the i-δth frame and marks it as L i-δ , based on the left camera image L i-δ and the right camera image R i Perform binocular vision synchronous positioning.

[0079] Step 11: Output binocular vision synchronous positioning results.

[0080] Step 12: i is incremented by 1, and the host computer 105 jumps to execute step 2.

[0081] In summary, the present invention uses a synchronous calibration device to generate calibration features, synchronously calibrates the left camera 101 and the right camera 102 to obtain the delay frame number δ, selects the left camera image or the right camera image of the current frame as the reference image based on the delay frame number δ, and searches for the synchronous frame image collected at the same time as the reference image. The host computer performs visual positioning based on the reference image and the synchronous frame image and outputs the target positioning result, which effectively improves the positioning accuracy. In addition, the present invention outputs the lag time t of binocular visual synchronous positioning while calibrating the binocular camera image. d , providing prior parameters for subsequent applications.

[0082] The above is only a preferred embodiment of the present application, and the present invention is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present invention should be considered to be included in the protection scope of the present invention.

Claims

1. A binocular vision synchronous positioning method, characterized in that: The binocular vision synchronous positioning method comprises: Provide a binocular camera and a host computer for controlling the working state of the binocular camera, wherein the binocular camera includes a left camera and a right camera; Based on the left camera image captured by the left camera and the right camera image captured by the right camera, the host computer synchronously calibrates the left camera and the right camera, so as to obtain the delay frame number δ representing the delay time when the left camera image and the right camera image are transmitted to the host computer after the synchronous calibration; During visual positioning, the host computer captures the left camera image and the right camera image of the current frame, selects the left camera image or the right camera image of the current frame as the reference image based on the delay frame number δ, and searches for the synchronous frame image acquired at the same time as the reference image. The host computer performs visual positioning based on the reference image and the synchronous frame image and outputs the target positioning result, wherein, When calibrating the left camera and the right camera synchronously, it includes: The host computer controls the synchronous calibration device to generate calibration features so that the calibration features fall into the field of view of the left camera and the right camera at the same time; When it is determined that the calibration features exist in both the left camera image and the right camera image, the host computer searches the collected left camera image and the right camera image to obtain the corresponding frame sequence m and frame sequence n when the calibration features first appear in the left camera image and the right camera image; Based on the frame sequence m and the frame sequence n, a delayed frame number δ is generated in the host computer, wherein the delayed frame number δ is the difference between the frame sequence m and the frame sequence n; If m≥n, the left camera image of the current i-th frame is selected as the reference image, and the i-δth frame is searched in the acquired right camera image as the synchronization frame image; If m<n, the right camera image of the current i-th frame is selected as the reference image, and the i-δth frame is searched in the acquired left camera image as the synchronization frame image.

2. The binocular vision synchronous positioning method according to claim 1, characterized in that: When performing binocular vision synchronous positioning, the host computer also outputs the lag time t d ,in, While controlling the synchronous calibration device to generate calibration features, the host computer records the initial time t of the lag calculation. o ; While outputting the target positioning result, the host computer records the lag calculation completion time t c ; The initial time t is calculated based on the hysteresis o and the hysteresis calculation is completed at time t c , host computer output lag time t d .

3. The binocular vision synchronous positioning method according to claim 2, characterized in that: When using binocular vision positioning, the host computer is based on the interval period T s The left camera and the right camera are calibrated synchronously, where: In each interval period T s After synchronous calibration, the current interval period T is obtained in the host computer. s The number of delayed frames δ.

4. The binocular vision synchronous positioning method according to claim 3, characterized in that: In the host computer, based on the lag time t d Adjust the interval period T s , then: Among them, T max is the upper limit of the interval period, T min is the lower limit of the interval period, t d [p] is the lag time of the pth output of the host computer, η is the lag time t d The fluctuation coefficient, N is the lag time t d The sample coefficient of is the lag time t d The sample mean of .

5. The binocular vision synchronous positioning method according to claim 4, characterized in that: Interval period T s is an integer, and the upper limit of the interval period is T max and the lower limit of the interval period T min Also an integer, where 100 <T min <200 and 300 <T max <400.

6. The binocular vision synchronous positioning method according to claim 4, characterized in that: During initialization, the host computer sets the upper limit of the interval period T max , Interval period lower limit T min , interval period T s and the initial value of the delay frame number δ, where T s =T min And δ=0.

7. A binocular vision synchronous positioning system, characterized in that: The binocular vision synchronous positioning system includes a synchronous calibration device for generating calibration features, a binocular camera for collecting image information, and a host computer for controlling the corresponding working states of the synchronous calibration device and the binocular camera, wherein: The host computer cooperates with the binocular camera and the synchronous calibration device to implement the binocular vision synchronous positioning method of any one of claims 1 to 6.

8. The binocular vision synchronous positioning system according to claim 7, characterized in that: The synchronous calibration device includes a laser, which is installed between the left eye camera and the right eye camera and generates laser light as the calibration feature. The laser light falls into the field of view of the left eye camera and the right eye camera at the same time.

9. The binocular vision synchronous positioning system according to claim 8, characterized in that: The binocular vision synchronous positioning system also includes a mechanical bracket, and the laser, the left-eye camera and the right-eye camera are fixed on the mechanical bracket.

Images