CN116543186A - Image matching method, system, device and medium in photoelectric navigation - Google Patents

Abstract

The invention discloses an image matching method, system, equipment and medium in photoelectric navigation, and relates to the technical field of photoelectric navigation; the method includes: performing correlation calculation based on the obtained compressed target frame at time t and the compressed reference frame at time t to obtain t The output vector at time, and then determine the compressed prediction vector at time t+1, and then judge whether it is in the set area, if not, replace the reference frame; if so, compare the compressed prediction vector at time t+1 with the compressed prediction vector at time t+1 Correlation calculation is performed between the target frame and the compressed reference frame at time t+1, and then the compressed motion vector at time t+1 is determined, and finally the final motion vector at time t+1 is determined to determine the moving route at time t+1, so that The target photoelectric navigation device moves according to the movement route at time t+1 at time t+1; the invention improves the calculation accuracy of the motion vector, can realize accurate image matching, makes the photoelectric navigation device move accurately according to the movement route, and improves the accuracy of motion tracking .

Description

Image matching method, system, device and medium in photoelectric navigation

technical field

The invention relates to the technical field of photoelectric navigation, in particular to an image matching method, system, equipment and medium in photoelectric navigation.

Background technique

In optoelectronic navigation technology, the commonly used motion estimation method is: first preset a range as the search range, then perform correlation calculations on the reference frame and the target frame, and find the search target block most correlated with the reference block as the optimal matching block. According to the relative position of the optimal matching block, the motion vector is output. Wherein, the preset interval is to determine a certain interval according to a certain predictive vector, and this interval should make the reference frame and the target frame have the strongest correlation (theoretically have overlapping image parts).

There may be a gap between the predicted vector and the actual motion vector. Image matching according to the motion estimation method in the prior art can correct the prediction vector within a certain range to obtain the actual motion vector, so that the images collected by the mouse within a certain speed range can be correctly matched. The existing motion estimation methods include full search, three-step search, four-step search, diamond search and other methods.

However, when the moving speed of the photoelectric navigation device changes greatly, the prediction vector cannot be correctly corrected according to the prior art. Taking the 3×3 neighborhood as an example to carry out the correlation calculation, the prediction vector can be corrected within ±1, that is, when the device moves, the speed change can be accurately calculated and tracked within 1. When the movement speed of the device changes by 2 or more, the prediction vector cannot be correctly corrected, so that the obtained current displacement is wrong. It will cause the matching to no longer locate the correct interval and find the optimal correlation module. The consequence is: the trajectory of the movement is completely abnormal. For example, in an optical mouse application, the mouse does not track motion correctly at some speeds, causing the cursor to fly around.

Contents of the invention

The object of the present invention is to provide an image matching method, system, equipment and medium in photoelectric navigation, by improving the calculation accuracy of motion vectors, accurate matching of images can be realized, so that photoelectric navigation equipment can move accurately according to the moving line, and improve the accuracy of motion tracking Spend.

To achieve the above object, the present invention provides the following scheme:

An image matching method in photoelectric navigation, the method comprising:

Obtain the electronic image collected by the target photoelectric navigation device at time t, and use the electronic image at time t as the target frame at time t; when t=1, the previous moment at time t is the initial time;

The image compression method is used to compress the target frame at time t to obtain the compressed target frame at time t;

Taking the electronic image at the initial moment as a reference frame at time t, and compressing the reference frame at time t to obtain a compressed reference frame at time t;

Perform correlation calculation according to the compressed reference frame at time t and the compressed target frame at time t to obtain an output vector at time t;

Determine the compressed predictive vector at t+1 time according to the output vector of the set number in t time;

Judging whether it is in the setting field according to the output vector at time t and the compressed prediction vector at time t+1;

If so, performing a correlation operation on the compressed reference frame at t+1 time and the compressed target frame at t+1 time according to the compressed prediction vector at t+1 time, to determine the compressed motion vector at t+1 time;

If not, use the target frame at time t as the reference frame at time t+1, and compress the reference frame at time t+1 to obtain the compressed reference frame at time t+1, and then return "According to the compression at time t The reference frame and the compressed target frame at time t carry out correlation calculation to obtain the output vector at time t" step;

Determine the final motion vector at time t+1 according to the compressed motion vector at time t+1; The mobile line moves at time t+1.

Optionally, the compressed prediction vector at time t+1 is determined according to a set number of output vectors within t time periods, specifically including:

When t>1, within t time, the average value of the set number of output vectors before t time is used as the compressed prediction vector at t+1 time, or the output vector at t time is used as the compressed prediction vector at t+1 time prediction vector;

When t=1, the output vector at time t is used as the compressed prediction vector at time t+1.

Optionally, the image compression method is an image compression method based on pixel replication, an image compression method based on interpolation, or an image compression method based on mean value merging.

Optionally, the final motion vector at time t+1 is determined according to the compressed motion vector at time t+1, specifically including:

Based on the image compression method, the compressed motion vector at time t+1 is decompressed to obtain the final motion vector at time t.

An image matching system in photoelectric navigation, said system comprising:

The image acquisition module is used to obtain the electronic image collected by the target photoelectric navigation device at time t, and the electronic image at time t is used as the target frame at time t; when t=1, the previous moment at time t is the initial time;

The image processing module is used to compress the target frame at time t by using image compression method to obtain the compressed target frame at time t;

A compression reference frame determination module is used to use the electronic image at the initial moment as a reference frame at time t, and compress the reference frame at time t to obtain a compressed reference frame at time t;

Calculation module, for performing correlation calculation according to the compression reference frame at time t and the compression target frame at time t, to obtain the output vector at time t;

Compressed predictive vector determination module, for determining the compressed predictive vector at t+1 moment according to the output vector of setting number in t time;

Judging module, for judging whether it is in the setting field according to the output vector at time t and the compressed prediction vector at time t+1;

The first determination module is used to perform a correlation operation on the compressed reference frame at time t+1 and the compressed target frame at time t+1 according to the compressed prediction vector at time t+1 when the result of the judgment module is yes, Determine the compressed motion vector at time t+1;

The second determining module is used for when the result of the judging module is no, then use the target frame at time t as the reference frame at time t+1, and compress the reference frame at time t+1 to obtain t+ The compressed reference frame at time 1, and then return to the "calculation module";

The motion vector determination module is used to determine the final motion vector at the time t+1 according to the compressed motion vector at the time t+1; the final motion vector at the time t+1 is used to determine the moving route at the time t+1, so that the target photoelectric navigation The device moves at time t+1 according to the moving route at time t+1.

Optionally, the compressed prediction vector determination module specifically includes:

The first determination sub-module is used for when t>1, within t moments, the average value of the output vectors set before the t moment is used as the compressed prediction vector at the t+1 moment, or the output at the t moment The vector is used as the compressed prediction vector at time t+1;

The second determination sub-module is configured to use the output vector at time t as the compressed prediction vector at time t+1 when t=1.

Optionally, the image processing module uses an image compression method based on pixel replication, an image compression method based on interpolation, or an image compression method based on mean value combination to perform compression processing.

Optionally, the motion vector determination module includes:

The determination sub-module is configured to decompress the compressed motion vector at time t+1 based on the image compression method to obtain the final motion vector at time t+1.

An electronic device includes a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the above-mentioned image matching method in optoelectronic navigation.

A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned image matching method in optoelectronic navigation is realized.

According to the specific embodiments provided by the invention, the invention discloses the following technical effects:

The present invention provides an image matching method, system, equipment and medium in photoelectric navigation. After compression processing is carried out by adopting an image compression method, a compressed reference frame and a compressed prediction vector are determined, and a motion vector is obtained after determining a compressed motion vector. Make the target photoelectric navigation device move according to the moving line provided by the motion vector; the present invention uses the image compression method to more accurately characterize the matching relationship between the predicted compressed target frame and the compressed reference frame, so that the calculated motion vector is more accurate , and then by improving the calculation accuracy of motion vectors, accurate image matching can be realized, so that the photoelectric navigation equipment can move accurately according to the moving route, and the accuracy of motion tracking can be improved.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some of the present invention. Embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without paying creative labor.

Fig. 1 is a flowchart of an image matching method in photoelectric navigation provided by an embodiment of the present invention;

Fig. 2 is a specific flow chart of the practical application of the image matching method in photoelectric navigation provided by the embodiment of the present invention;

FIG. 3 is a schematic diagram of a search area provided by an embodiment of the present invention;

FIG. 4 is a schematic diagram of Example 1 of a matching correlation operation in the prior art provided by an embodiment of the present invention;

FIG. 5 is a schematic diagram of Example 2 of matching correlation calculation in the prior art provided by an embodiment of the present invention;

FIG. 6 is a schematic diagram of Example 3 of a matching correlation operation in the prior art provided by an embodiment of the present invention;

FIG. 7 is a schematic diagram of Example 4 of matching correlation calculation in the prior art provided by an embodiment of the present invention;

FIG. 8 is a schematic diagram of Example 5 of matching correlation operations in the prior art provided by an embodiment of the present invention;

Fig. 9 is a structural diagram of an image matching system in photoelectric navigation provided by an embodiment of the present invention.

Symbol Description:

Image acquisition module-1, image processing module-2, compression reference frame determination module-3, calculation module-4, compression prediction vector determination module-5, judgment module-6, first determination module-7, second determination module- 8. Motion vector determination module-9.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

The object of the present invention is to provide an image matching method, system, equipment and medium in photoelectric navigation, by improving the calculation accuracy of motion vectors, accurate matching of images can be realized, so that photoelectric navigation equipment can move accurately according to the moving line, and improve the accuracy of motion tracking Spend.

In order to make the above objects, features and advantages of the present invention more comprehensible, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

When the LED lighting system irradiates the target plane, the reflected light of the target plane enters the device sensor to form an electronic image, and the device processes the electronic image (such as image compression, image matching) to obtain the motion vector. This type of device is called photoelectric navigation device.

In the prior art, when the moving speed of the photoelectric navigation device changes greatly, that is, when the movement amount of two adjacent frames of images changes greatly, the current displacement obtained according to the predicted vector is wrong, which will cause the matching to fail again. Unable to locate the correct interval and find the optimal correlation module. For example, in an optical mouse application:

The mouse doesn't track movement properly at some speeds and the cursor flies around.

The technical problem to be solved by the present invention is to correctly calculate the current motion vector when the moving speed of the photoelectric navigation device changes greatly.

In the general process, that is, the motion estimation algorithm, the present invention adds a function of compressing large array images into small array images, and matching the compressed images, that is, compressing target frames synchronized with the general process, and performing image matching to obtain accurate motion vector (dx, dy). To ensure that the photoelectric navigation equipment can correctly calculate the motion vector when the moving speed changes greatly.

Among them, the large array image collected by the image sensor is compressed into a small array image, which is referred to as compressed array for short. This process preserves the morphology and important features of the original image while compressing it.

Image matching means that the image sensor captures the first frame image as a reference image for the first time, that is, the reference frame, and captures the second frame image as the target image for the second time, that is, the target frame; comparing the reference frame with the target frame, the reference frame and the target frame Motion estimation is performed between frames to determine the moving direction and distance of the target frame relative to the reference frame, that is, the motion vector.

Motion estimation method: preset a reference block from the reference frame, select a block as the starting point in the target frame according to a principle, and search for the target block starting from a section centered on the starting point, and search for the target block in a search range by Gradually expand the search from inside to outside to obtain the search target block most correlated with the reference block as the optimal matching block, so as to determine the moving direction and distance (motion vector) of the target frame relative to the reference frame.

Example 1

As shown in Figure 1, an embodiment of the present invention provides an image matching method in photoelectric navigation, the method includes:

Step 100: Obtain the electronic image collected by the target photoelectric navigation device at time t, and use the electronic image at time t as the target frame at time t; when t=1, the previous time at time t is the initial time.

Step 200: Compress the target frame at time t by using an image compression method to obtain the compressed target frame at time t. Specifically, the image compression method is an image compression method based on pixel replication, an image compression method based on interpolation, or an image compression method based on mean value merging.

Step 300: Use the electronic image at the initial time as a reference frame at time t, and compress the reference frame at time t to obtain a compressed reference frame at time t.

Step 400: Perform correlation calculation according to the compressed reference frame at time t and the compressed target frame at time t to obtain an output vector at time t.

Step 500: Determine the compressed prediction vector at time t+1 according to the set number of output vectors within t time periods.

Wherein, according to the set number of output vectors in the t time period, the compressed prediction vector at the time t+1 is determined, specifically including:

When t>1, within t time, the average value of the set number of output vectors before t time is used as the compressed prediction vector at t+1 time, or the output vector at t time is used as the compressed prediction vector at t+1 time prediction vector.

When t=1, the output vector at time t is used as the compressed prediction vector at time t+1.

In short, the compressed prediction vector can be the output vector at the previous moment, or the linear relationship of the output vectors at several previous moments. That is, several output vectors are selected from the output vectors at several moments, and linear operations are performed on them.

Step 600: According to the output vector at time t and the compressed prediction vector at time t+1, it is judged whether it is in the set area.

Being in the setting field means that the compression reference frame and the compression target frame are highly correlated, and there is no need to replace the reference frame; if it is not in the setting field, the reference frame needs to be replaced.

Step 700: If yes, perform a correlation operation on the compressed reference frame at time t+1 and the compressed target frame at time t+1 according to the compressed prediction vector at time t+1, and determine the compressed motion vector at time t+1.

Step 800: If not, use the target frame at time t as the reference frame at time t+1, and compress the reference frame at time t+1 to obtain a compressed reference frame at time t+1, and then return to step 400.

Step 900: Determine the final motion vector at time t+1 according to the compressed motion vector at time t+1; the final motion vector at time t+1 is used to determine the moving route at time t+1, so that the target photoelectric navigation device can Time 1 moves according to the movement route at time t+1.

Wherein, step 700 specifically includes:

Based on the image compression method, the compressed motion vector at time t+1 is decompressed to obtain the final motion vector at time t+1.

Fig. 2 is a specific flow chart of the practical application of the image matching method in photoelectric navigation provided by the embodiment of the present invention. In practical application, the concrete flow process of the present invention is as follows:

1. Obtain an electronic image collected by the target photoelectric navigation equipment as a reference frame, and obtain another electronic image as a target frame; that is, obtain one image as a reference frame ref and the subsequent image as a target frame tar, wherein the reference frame meets the requirements of After the condition needs to be replaced, the target frame needs to be acquired and updated every time.

Among them, the conditions to be met are: the correlation operation is performed in the search field, and the search field is small, so the correlation operation result has locality and limitations; when the search field is smaller than a certain threshold, it is judged that it is not suitable for the correlation operation , the reference frame needs to be replaced.

2. Compress the reference frame and the target frame to obtain the compressed reference frame ref_check and the compressed target frame tar_check, wherein the compressed reference frame and the compressed target frame are acquired and updated each time. Compression methods such as image compression based on pixel copy method, image compression based on interpolation, or image compression that takes the mean value of small squares such as 2×2 and 3×3 in the original square matrix and merges them into one block.

3. According to the compressed prediction vector (predx_check, predy_check) and the actual motion vector of the last compressed target frame and the compressed reference frame, image matching is performed on the compressed reference frame and the compressed target frame to obtain the compressed motion vector (dx_check, dy_check). The compressed prediction vector can be the output vector at the previous moment, or the linear relationship of the output vectors at several previous moments. The linear relationship may be the mean value of the previous output values or others, which is not limited here.

4. Output the motion vector (dx, dy) according to the compressed motion vector. Motion vectors can be output according to different compression methods. If compressed at a ratio of 2:1, the output is output at a ratio of 1:2.

According to the position of the compression target frame at the previous moment and the compression prediction vector at the current moment, the position of the compression target frame at the current moment can be predicted, that is, according to the actual motion vector of the compression target frame at the previous moment and the corresponding compression reference frame and the compression at the current moment The prediction vector can obtain the compression motion vector between the compression target frame at the current moment and the compression reference frame at the current moment. Referring to FIG. 3 , the overlapping area between the reference frame at the current moment and the target frame at the current moment is determined as the search area (ie, B in FIG. 3 ). The search area is the setting area mentioned above. This region makes the reference frame at the current moment have the strongest correlation with the target frame at the current moment. Determine the reference block and the target block in the search area, perform a correlation operation, and modify the prediction vector to obtain the correct motion vector (dx, dy) of the current target frame relative to the previous target frame.

5. When performing the next image matching, it is necessary to consider whether to change the reference frame first. When the search area is too small, the correlation between the reference frame and the target frame is not strong, which may cause errors in matching and prediction, so the current target frame needs to be used as the reference frame.

6. The existing scheme takes a 3×3 neighborhood as an example to perform correlation calculations, such as the examples of correlation calculations performed on nine matching target blocks and reference blocks, see Figure 4 to Figure 8 for details; the prediction vector can be ±1 The correction within , when the device moves, the speed change can be accurately tracked and calculated within 1 of the motion vector.

The correction of the predictive vector is to determine a fixed reference block in the reference frame and an initial search target block in the target frame according to the predictive vector, and use the center of the initial search target block as the center of the 3×3 neighborhood to move and match around, and get the same as The most relevant search target block of the reference block is used as the optimal matching block, and the prediction vector is corrected according to the relative position of the optimal matching block and the initial search target block.

Specifically, the reference block and the search target block are determined, and a correlation operation is performed between the reference block and multiple designated search target blocks in the neighborhood. Taking a 3×3 neighborhood as an example, the correlation results can be recorded as 0-9. According to the compressed prediction vector, and compare 0 to 9, find the optimal correlation value, the target block corresponding to the optimal correlation value, that is, the relative movement position between the optimal matching block and the reference block, and move it to the relative position to compress the prediction After the vector is corrected, it is used as the motion vector value (dx, dy) of this displacement. The target block is a block that overlaps with the reference block and its neighbor blocks within the search area on the target frame.

In the present invention, when the moving speed of the photoelectric navigation device changes, with regard to Fig. 4, the search target block centered on the "0" point and the reference target block perform correlation calculations; then the search target block moves one grid to the upper left corner, Take point "1" as the center, see Figure 5. Next, the search target block is moved to the upper corner by one grid, centered on point "3", see Fig. 6 . Repeating this process, the actual most relevant block finally obtained is shown in Figure 8. Wherein, the symbol * in FIG. 8 represents the center point in the initial search target block, that is, the position centered on point “0” in FIG. 4 . In Figure 8, the most relevant matching block center points are outside the 3×3 neighborhood. That is, when the movement speed of the photoelectric navigation device changes by 2 or more, the center point of the most relevant matching block exceeds the 3×3 neighborhood, resulting in the inability to locate the correct interval and making the motion trajectory of the photoelectric navigation device abnormal.

After the compression algorithm is adopted in the embodiment of the present invention, it can also be accurately tracked when the moving speed of the photoelectric navigation device changes greatly.

Wherein, in the above four, a predictive vector may also be obtained according to the compressed motion vector, and image matching is performed on the uncompressed image again to output the motion vector. It is not limited to this process and does not affect the concept of the present invention.

In addition, the correlation operation mentioned in the embodiment of the present invention refers to the judgment of the similarity of the pixels corresponding to the reference block and the target block.

Common grayscale-based correlation algorithms include: MedianAbsolute Deviation (MAD), sum of absolute errors (SumofAbsoluteDifferences, SAD), fully convolutional network detection (SingleShotMultiBoxDetector, SSD), etc.

In the SAD algorithm, two matching blocks (reference block, target block) are subtracted point-to-point, then the absolute value is taken, and finally all absolute values are summed. If the target block is exactly the same as the reference block, then the correlation is optimal and its SAD value is 0.

Example 2

As shown in Figure 9, in order to implement the method corresponding to the above-mentioned embodiment 1 to achieve corresponding functions and technical effects, the embodiment of the present invention provides an image matching system in photoelectric navigation, the system includes: an image acquisition module 1, an image Processing module 2 , compression reference frame determination module 3 , calculation module 4 , compression prediction vector determination module 5 , judgment module 6 , first determination module 7 , second determination module 8 and motion vector determination module 9 .

The image acquisition module 1 is used to acquire the electronic image collected by the target photoelectric navigation device at time t, and use the electronic image at time t as the target frame at time t; when t=1, the previous time at time t is the initial time.

The image processing module 2 is configured to compress the target frame at time t by using an image compression method to obtain the compressed target frame at time t. Wherein, the image processing module 2 adopts an image compression method based on pixel duplication, an image compression method based on interpolation, or an image compression method based on mean value combination to perform compression processing.

The compressed reference frame determining module 3 is configured to use the electronic image at the initial time as the reference frame at the time t, and compress the reference frame at the time t to obtain the compressed reference frame at the time t.

The calculation module 4 is configured to perform correlation calculation according to the compressed reference frame at time t and the compressed target frame at time t to obtain an output vector at time t.

The compressed predictive vector determining module 5 is configured to determine the compressed predictive vector at time t+1 according to a set number of output vectors within t time periods.

Wherein, the compression prediction vector determining module 5 specifically includes: a first determining sub-module and a second determining sub-module.

The first determination sub-module is used for when t>1, within t moments, the average value of the output vectors set before the t moment is used as the compressed prediction vector at the t+1 moment, or the output at the t moment The vector is used as the compressed prediction vector at time t+1.

The second determination sub-module is configured to use the output vector at time t as the compressed prediction vector at time t+1 when t=1.

The judging module 6 is used for judging whether the output vector at time t and the compressed prediction vector at time t+1 is in the set area.

The first determination module 7 is used to perform a correlation operation on the compressed reference frame at time t+1 and the compressed target frame at time t+1 according to the compressed prediction vector at time t+1 when the result of the judgment module is yes. , to determine the compressed motion vector at time t+1.

The second determination module 8 is used for when the result of the judgment module is no, then use the target frame at time t as the reference frame at time t+1, and compress the reference frame at time t+1 to obtain t +1 the compressed reference frame at the moment, and then return to "computation module 4".

Motion vector determination module 9 is used to determine the final motion vector at t+1 moment according to the compressed motion vector at t+1 moment; the final motion vector at t+1 moment is used to determine the moving line at t+1 moment, so that the target photoelectric The navigation device moves at time t+1 according to the moving route at time t+1.

Specifically, the motion vector determination module 9 includes: a determination sub-module.

The determination sub-module is configured to decompress the compressed motion vector at time t+1 based on the image compression method to obtain the final motion vector at time t+1.

Example 3

An electronic device includes a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the image matching method in photoelectric navigation in Embodiment 1.

As an optional implementation manner, the foregoing electronic device may be a server.

In one embodiment, the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the image matching method in the optoelectronic navigation in Embodiment 1 is realized.

Beneficial effects of the present invention:

1. When the moving speed of the photoelectric navigation equipment changes greatly, the movement amount between two consecutive frames of images collected by the image sensor changes relatively greatly. The compression algorithm is more accurate than the algorithm used in the prior art to preset the reference frame and The target frame has the strongest correlation matching interval, and the current motion vector is correctly calculated.

2. The existing technology takes a 3x3 neighborhood as an example to perform correlation calculations, and can correct the prediction vector within ±1, that is, when the device moves, the speed change can be accurately calculated and tracked within 1. When the movement speed of the device changes by 2 or more, the prediction vector cannot be corrected correctly.

After using the compression algorithm, taking 1/2 compression of the original square matrix as an example, the compressed image is correlated with a 3×3 neighborhood, and the correction of ±1 to the compressed prediction vector is equivalent to performing a correction on the prediction vector of the prior art The correction of ±2 means that the motion vector when the moving speed of the device changes to 2 can be accurately calculated.

3. When the collected image is relatively flat and blurred, the image matching in the prior art is easy to make mistakes, and the correct rate of the image matching result is improved after the image is compressed.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other.

In this paper, specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; meanwhile, for those of ordinary skill in the art, according to the present invention Thoughts, there will be changes in specific implementation methods and application ranges. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims (10)

1. An image matching method in photoelectric navigation, the method comprising:

acquiring an electronic image acquired by target photoelectric navigation equipment at the time t, and taking the electronic image at the time t as a target frame at the time t; when t=1, the time immediately preceding the time t is the initial time;

performing compression processing on the target frame at the time t by adopting an image compression method to obtain a compressed target frame at the time t;

taking the electronic image at the initial moment as a reference frame at the t moment, and compressing the reference frame at the t moment to obtain a compressed reference frame at the t moment;

performing correlation calculation according to the compressed reference frame at the time t and the compressed target frame at the time t to obtain an output vector at the time t;

determining a compression prediction vector at the time t+1 according to the set number of output vectors in the time t;

judging whether the set field exists or not according to the output vector at the time t and the compression prediction vector at the time t+1;

if so, carrying out correlation operation on the compression reference frame at the time t+1 and the compression target frame at the time t+1 according to the compression prediction vector at the time t+1, and determining a compression motion vector at the time t+1;

if not, taking the target frame at the time t as a reference frame at the time t+1, compressing the reference frame at the time t+1 to obtain a compressed reference frame at the time t+1, and returning to the step of carrying out correlation calculation according to the compressed reference frame at the time t and the compressed target frame at the time t to obtain an output vector at the time t;

determining a final motion vector at the time t+1 according to the compressed motion vector at the time t+1; the final motion vector at the time t+1 is used for determining a moving line at the time t+1 so that the target photoelectric navigation device moves at the time t+1 according to the moving line at the time t+1.

2. The method for matching images in electro-optical navigation according to claim 1, wherein determining the compression prediction vector at time t+1 according to the set number of output vectors at time t comprises:

when t >1, taking the average value of the output vectors of the preset number before t as the compression prediction vector of t+1 time or taking the output vector of t as the compression prediction vector of t+1 time in t times;

when t=1, the output vector at time t is taken as the compression prediction vector at time t+1.

3. The image matching method in electro-optical navigation according to claim 1, wherein the image compression method is an image compression method based on a pixel replication method, an image compression method based on interpolation, or an image compression method of mean combination.

4. The method for matching images in electro-optical navigation according to claim 1, wherein determining a final motion vector at time t+1 from the compressed motion vector at time t+1, specifically comprises:

and (3) based on the image compression method, decompressing the compressed motion vector at the time t+1 to obtain the final motion vector at the time t.

5. An image matching system in electro-optical navigation, the system comprising:

the image acquisition module is used for acquiring an electronic image acquired by the target photoelectric navigation equipment at the time t and taking the electronic image at the time t as a target frame at the time t; when t=1, the time immediately preceding the time t is the initial time;

the image processing module is used for compressing the target frame at the moment t by adopting an image compression method to obtain a compressed target frame at the moment t;

the compression reference frame determining module is used for taking the electronic image at the initial moment as a reference frame at the t moment, and carrying out compression processing on the reference frame at the t moment to obtain a compression reference frame at the t moment;

the calculation module is used for carrying out correlation calculation according to the compressed reference frame at the time t and the compressed target frame at the time t to obtain an output vector at the time t;

the compression prediction vector determining module is used for determining a compression prediction vector at the time t+1 according to the set number of output vectors in the time t;

the judging module is used for judging whether the set field exists or not according to the output vector at the time t and the compression prediction vector at the time t+1;

the first determining module is used for performing correlation operation on the compression reference frame at the time t+1 and the compression target frame at the time t+1 according to the compression prediction vector at the time t+1 when the result of the judging module is yes, and determining the compression motion vector at the time t+1;

the second determining module is used for taking the target frame at the time t as the reference frame at the time t+1 and carrying out compression processing on the reference frame at the time t+1 to obtain a compressed reference frame at the time t+1 when the result of the judging module is negative, and then returning to the calculating module;

the motion vector determining module is used for determining a final motion vector at the time t+1 according to the compressed motion vector at the time t+1; the final motion vector at the time t+1 is used for determining a moving line at the time t+1 so that the target photoelectric navigation device moves at the time t+1 according to the moving line at the time t+1.

6. The system for matching images in electro-optical navigation according to claim 5, wherein the compression prediction vector determining module specifically comprises:

a first determination submodule, configured to, when t >1, set, in t times, an average value of a number of output vectors before t times as a compression prediction vector at t+1 times, or set, in t times, an output vector at t times as a compression prediction vector at t+1 times;

and the second determination submodule is used for taking the output vector at the time t as the compression prediction vector at the time t+1 when t=1.

7. The image matching system in the electro-optical navigation according to claim 5, wherein the image processing module performs the compression processing by using an image compression method based on a pixel replication method, an image compression method based on interpolation, or an image compression method based on a mean combination.

8. The electro-optical navigation image matching system of claim 5, wherein the motion vector determination module comprises:

and the determining submodule is used for decompressing the compressed motion vector at the time t+1 based on the image compression method to obtain the final motion vector at the time t+1.

9. An electronic device comprising a memory for storing a computer program and a processor that runs the computer program to cause the electronic device to perform the method of image matching in electro-optical navigation according to any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by a processor, implements the image matching method in electro-optical navigation according to any one of claims 1 to 4.