CN114937076A - A dimension measuring method and device - Google Patents

Abstract

The present application provides a size measurement device and a size measurement method. The size measurement device includes: a depth detection module with a pixel array for acquiring depth data within a field of view; an aiming module for aiming at the A target item within the field of view of the depth detection module, wherein the aiming pattern is pre-calibrated corresponding to the pixel positions in the pixel array; a memory pre-stores the pixel positions corresponding to the aiming pattern; a processor, It is used to convert the depth data obtained by the depth detection module into a depth point cloud, and calculate the size of the target item according to the depth point cloud corresponding to the pixel position. It enables the processor to calculate the size of the target item according to the depth point cloud obtained by the depth detection module, and will not be disturbed by nearby items.

Description

A dimension measuring method and device

technical field

The present application relates to the field of dimension measurement, in particular to a dimension measurement method and device.

Background technique

The accuracy of the size of the goods has positive significance in the mail charging and the allocation of storage space. Traditionally, the size of the goods can be measured manually and input into the computer manually. With the development of technology, a variety of automatic ranging technologies have been gradually developed. , such as structured light ranging and binocular ranging, etc., based on these automatic ranging technologies, a variety of volume measurement technologies have been developed. At present, these volume measurement technologies are relatively mature in measuring the volume of individual goods and have high measurement accuracy, such as measuring The volume of a single small cargo placed on open ground can be very precise.

However, when the goods are very dense, the goods next to it will interfere with the goods to be measured, and the user needs to move the goods to be measured to an open location for measurement, which is laborious and inconvenient to operate.

In view of the above problems, the present application provides a new dimension measurement method and device, and adopts new methods and technical means to solve these problems.

SUMMARY OF THE INVENTION

The purpose of the creation of the present application is to provide a size measurement method and device that can aim at the object to be targeted to eliminate interference.

To achieve the above object, the application adopts the following technical means:

The application provides a size measurement device, which is characterized by comprising: a depth detection module with a pixel array for acquiring depth data within a field of view; an aiming module for aiming at the depth detection module through an aiming pattern A target object within a group of fields of view, wherein the aiming pattern is pre-calibrated corresponding to the pixel positions in the pixel array; a memory, pre-stores the pixel positions corresponding to the aiming pattern; a processor is used to The depth data obtained by the depth detection module is converted into a depth point cloud, and the size of the target item is calculated according to the depth point cloud corresponding to the pixel position.

Optionally, the aiming module includes a display screen, and an aiming pattern is displayed at a fixed position of the display screen, and the display screen is used to display the image of the target item. When the image of the target item is overlapped, the image of the target item is at least partially imaged at the pixel position corresponding to the aiming pattern.

Optionally, a sizing application is included to generate the aiming pattern in an application interface.

Optionally, a display screen is included, and the processor is configured to calculate the vertex position of the target item according to an algorithm, and generate the cuboid envelope of the target item on the display screen in real time based on the vertex position of the target item.

Optionally, when there are multiple items in the field of view of the depth detection module, the processor cuts and fits the depth point cloud into multiple point cloud planes, and the point cloud plane overlaps with the aiming pattern. The corresponding item is the target item.

The present application provides a size measurement method, which is characterized by comprising: a size measurement device comprising: a depth detection module, which acquires depth data within a field of view by having a pixel array; the aiming module generates an aiming pattern to aim at the depth Detecting target objects within the field of view of the module, and the aiming pattern is pre-calibrated corresponding to the pixel positions in the pixel array; the memory pre-stores the pixel positions corresponding to the aiming pattern; the processor stores the pixel positions corresponding to the aiming pattern in advance. The depth data acquired by the depth detection module is converted into a depth point cloud, and the size of the target item is calculated according to the depth point cloud corresponding to the pixel position.

Optionally, the aiming module includes a display screen, and a fixed position of the display screen displays an aiming pattern, the display screen is used to display the image of the target item, and aiming at the target item through the aiming pattern includes: adjusting The orientation of the sizing device so that the aiming pattern overlaps the image of the target item on the display screen.

Optionally, it also includes: generating the aiming pattern through a size measurement application.

Optionally, a display screen is further included, wherein the processor calculates the vertex position of the target item in real time, and generates the cuboid envelope of the target item on the display screen in real time based on the vertex position of the target item.

Optionally, when there are multiple items in the field of view of the depth detection module, the processor cuts and fits the depth point cloud into multiple point cloud planes, and the point cloud plane overlaps with the aiming pattern. The corresponding item is the target item.

Compared with the prior art, the present application has the following beneficial effects:

According to the size measurement method and device of the present application, the size measurement device aims at the target item within the field of view of the depth detection module through the aiming pattern, and plays a guiding role, and the aiming pattern corresponds to the pixels in the pixel array of the depth detection module The position is pre-calibrated and stored in the memory, so that the processor calculates the size of the target item according to the depth point cloud obtained by the depth detection module, and will not be disturbed by nearby items.

Description of drawings

FIG. 1 is a schematic diagram of a size measurement device measuring size according to an embodiment of the application;

Fig. 2 is the structural block diagram of the dimension measuring device in Fig. 1;

Fig. 3 is the partial schematic diagram of the application program interface of the size measurement application program on the size measurement device in Fig. 1;

4 is a schematic diagram of displaying a target item image on the application program interface in FIG. 3;

FIG. 5 is a flow chart of a method for measuring the volume of the volume measuring device in FIG. 1 .

Description of the reference numerals for the specific embodiments:

Dimension measuring device 100; processor 101; depth detection module 102; camera 103; display screen 104; memory 105; aiming pattern 106; cuboid envelope 107; target object 200;

Detailed ways

In order to facilitate a better understanding of the purpose, structure, features and effects of the present application, the present application will now be further described with reference to the accompanying drawings and specific embodiments.

The present application discloses a size measurement method and device, which are not only suitable for the size measurement of items packaged in a regular cuboid, but also for the size measurement of irregular items. The size measurement device 100 of the present application may be a handheld size measurement device or a size measurement robot.

An embodiment as shown in FIG. 1 shows a situation in which the hand-held size measuring device 100 collects data, and a user holds the size measuring device 100 to measure the size of an item.

Supplementary reference is made to the structural block diagram of the size measurement apparatus 100 of an embodiment shown in FIG. 2 . The size measurement device 100 mainly includes: a processor 101 , a depth detection module 102 , a camera 103 , a display screen 104 and a memory 105 . In this embodiment, the camera 103 and the display screen 104 are used as part of the aiming module; in some embodiments (not shown, the same below), the aiming module may be optical aiming, such as through laser or LED projection Aiming pattern 106 aids in aiming and the like.

While the aiming module is aiming, the depth detection module 102 starts to collect depth data, and the processor 101 converts the depth data into a point cloud in the background. The depth detection module 102 may be a time-of-flight camera, an active stereo camera, One or a combination of passive stereo vision cameras, structured light cameras, and stereo image cameras. The processor 101 is configured to convert the depth data collected by the depth detection module 102 into a depth point cloud, cut the depth point cloud into a plurality of blocks, and expand and fit the same blocks into a plane through a clustering algorithm, and then According to the correlation between the planes, the size of the item is obtained. For example, if two planes with a common edge belong to the same item, the length, width and height of the item are obtained according to the two planes with a common edge, and then the volume of the item is calculated; Combine the top surface (to get the length and width of the item) and the ground, and the distance between the top and the ground (the height of the item) to obtain the length, width and height of the item, and then calculate the volume of the item. The memory 105 stores a dimension measurement application, and the processor 101 can run the dimension measurement application.

FIG. 3 and FIG. 4 show a part of the application program interface of the size measurement application according to an embodiment, the application program interface will automatically generate the aiming pattern 106, when measuring the size, first open the size measurement application program , at the same time, the camera 103 starts to collect images, the display screen 104 displays the images collected by the camera 103, the depth detection module 102 starts to collect depth data within the field of view, wherein the aiming pattern 106 It is located at a fixed position on the display screen 104, and the aiming pattern 106 is pre-calibrated and stored in the memory 105 corresponding to the pixel positions in the pixel array of the depth detection module 102. The depth detection module 102 is also pre-calibrated and aligned, so that when the aiming pattern 106 is aimed at the two-dimensional image of the target item 200 collected by the camera 103 in the display screen 104, all the depth detection module 102 The pixel position just collects the depth data of the target item 200, the processor 101 converts the depth data collected by the depth detection module 102 into a depth point cloud, and calculates the depth point cloud according to the depth point cloud corresponding to the aiming pattern 106. Describe the size of the target item 200. When optical aiming is employed, the optical aiming pattern 106 is also pre-calibrated and stored corresponding to the pixel positions in the pixel array.

Specifically, continuing to refer to FIG. 4 , the two-dimensional image captured by the camera 103 is displayed on the display screen 104 , and the user adjusts the posture of the size measurement device 100 so that the aiming pattern 106 is aligned with the target item on the display screen 104 . When one surface (such as the top surface) of 200 is overlapped, at the same time, the processor 101 converts the depth data obtained by the depth detection module 102 into a depth point cloud, and fits a plurality of planes according to the depth point cloud, which are compared with the depth point cloud. The plane where the depth point cloud collected by the pixel position corresponding to the aiming pattern 106 is located is a surface of the target item 200, the vertex position of the surface is obtained, the length, width and height of the target item 200 are obtained according to the aforementioned algorithm, and based on the target The vertex position and length, width and height information of the item 200 are used to generate the cuboid envelope 107 of the target item 200 in real time. The length, width and height of the envelope 107 correspond to the length, width and height of the target item 200 in the image, which is convenient for the user to check whether the size measurement is correct. The volume of the target object 200 is large, and when the length, width and height of the cuboid envelope 107 and the target object 200 have a large deviation, the user can adjust the posture of the size measurement device 100 in time, re-target the target object 200 and re-measure the size of the target object 200 . At the same time, when the target object 200 is not aimed, the processor 101 cannot obtain the size of the target object 200, so that the cuboid envelope 107 will not be generated. For example, a cuboid envelope 107 is generated on the targeted object 200 in FIG. 4 , and another object above the target object 200 is not targeted, so the cuboid envelope 107 is not generated on the object. When there are multiple objects in the field of view of the depth detection module 102, the processor 101 will only calculate the size of the target object 200 targeted by the aiming pattern 106, and ignore other objects next to it, so that only the target object The cuboid envelope 107 is generated on 200 , indicating that the user has correctly aimed at the target item 200 .

In this embodiment, in order to facilitate the user to view the image 201 of the target item, the camera 103 is used to collect the two-dimensional image of the target item 200, and the two-dimensional image is displayed on the display screen 104. The group 102 is calibrated so that when the aiming pattern 106 is aimed at the two-dimensional image of the target item 200 collected by the camera 103 in the display screen 104, the pixels of the depth detection module 102 corresponding to the aiming pattern 106 The position also just collects the depth data of the target item 200 . In some embodiments, images are not captured by the camera 103, but only depth data is captured by the depth detection module 102. The depth data is then converted into a depth image and displayed on the display screen 104, while the aiming pattern is used to capture depth data. 106 Aiming at the target item 200 in the depth image can also have the effect of aiming, and only the pixel position of the depth detection module 102 corresponding to the aiming pattern 106 needs to be calibrated, without the need for the camera 103 and the pixel position. The depth detection module 102 performs calibration and alignment.

FIG. 5 is a flow chart of the method for measuring the size of the size measuring device 100 according to the aforementioned embodiment, and the details are as follows:

S1: The aiming module generates the aiming pattern 106 .

Specifically, the user opens a sizing application on the display screen 104, and a fixed position of its application interface generates the aiming pattern 106.

S2: The depth detection module 102 acquires depth data within the field of view.

The aiming pattern 106 is pre-calibrated and stored in the memory 105 corresponding to the pixel positions in the pixel array of the depth detection module 102 .

S3: The processor 101 converts the depth data obtained by the depth detection module 102 into a depth point cloud, and calculates the size of the target item 200 according to the depth point cloud corresponding to the pixel position.

The processor 101 synchronously converts the depth data into a depth point cloud/depth image and displays it on the display screen 104, and when the user adjusts the aiming pattern 106 to aim at the target object 200 in the depth point cloud/depth image, the aiming pattern 106 and the When the images 201 of the target item overlap, the processor 101 determines that the item targeted by the aiming pattern 106 is the target item 200, and calculates dimensions such as length, width, height or volume of the target item 200 according to the depth point cloud.

On the other hand, two-dimensional graphics can also be collected through the camera 103, and the two-dimensional graphics collected by the camera 103 or the depth point cloud/depth image collected by the depth detection module 102 can be switchably displayed through the display screen 104, and the aiming pattern 106 and the camera 103 and the depth detection module 102 are both calibrated, so that the aiming pattern 106 overlaps with the target item 200 in the two-dimensional graphics or overlaps with the target item 200 in the depth point cloud/depth image, indicating that the aiming pattern 106 is aimed at the target item 200. , the processor 101 will calculate the vertex positions of the target item 200 and generate the cuboid envelope 107 in real time. The specific calculation details have been described above and will not be repeated here.

It should be understood that the size of the sequence numbers of the steps in the above embodiments does not mean the sequence of execution, and the execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application. For example, S1 and S2 may be performed sequentially or simultaneously, or S2 may be performed first, and then S1 may be performed after a certain condition is satisfied (eg, controlled by a user).

On the other hand, the aforementioned size measurement device 100 and its size measurement method are also applicable to the size measurement of irregular items. When the user aims at the target irregular item through the aiming pattern 106, the processor 101 automatically generates the minimum size of the target irregular item according to the algorithm. Circumscribe the cube to obtain the dimensions such as the length, width, height or volume of the target irregular item.

A dimension measurement method and device of the present application have the following beneficial effects:

In the size measurement method and device of the present application, the size measurement device 100 aims at the target object 200 within the field of view of the depth detection module 102 through the aiming pattern 106 to play a guiding role, and the aiming pattern 106 corresponds to the depth detection module 102 The pixel positions in the pixel array are pre-calibrated and stored in the memory 105, so that the processor 101 calculates the size of the target object 200 according to the depth point cloud obtained by the depth detection module 102, without being disturbed by nearby objects.

The above detailed description is only the description of the preferred embodiment of the present application, and is not intended to limit the scope of the patent of the present application. Therefore, any equivalent technical changes made by using the contents of the description and illustrations of this creation are included in the patent of this creation. within the range.

Claims (10)

1. A dimensional measurement device, comprising:

the depth detection module is provided with a pixel array and is used for acquiring depth data in a field range;

the aiming module is used for aiming a target object within the field of view of the depth detection module through an aiming pattern, wherein the aiming pattern is calibrated in advance corresponding to the pixel position in the pixel array;

a memory for pre-storing the pixel position corresponding to the aiming pattern;

and the processor is used for converting the depth data acquired by the depth detection module into a depth point cloud and calculating the size of the target object according to the depth point cloud corresponding to the pixel position.

2. The dimensional measurement device of claim 1, wherein: the aiming module comprises a display screen, an aiming pattern is displayed at a fixed position of the display screen, the display screen is used for displaying an image of the target object, and when the aiming pattern is overlapped with the image of the target object on the display screen, at least part of the image of the target object is imaged at the pixel position corresponding to the aiming pattern.

3. The dimensional measurement device of claim 1, wherein: a sizing application is included to generate the aiming pattern at an application interface.

4. The dimensional measurement device of claim 1, wherein: the processor is used for calculating the vertex position of the target object according to an algorithm and generating a cuboid envelope of the target object on the display screen in real time based on the vertex position of the target object.

5. The dimensional measurement device of claim 1, wherein: when a plurality of articles exist in the field range of the depth detection module, the processor cuts and fits the depth point cloud into a plurality of point cloud planes, and the articles corresponding to the point cloud planes overlapped with the aiming patterns are target articles.

6. A dimensional measurement method, comprising:

the size measuring device comprises: the depth detection module acquires depth data in a field range through a pixel array; the aiming module generates an aiming pattern to aim at the target object within the field of view of the depth detection module, and the aiming pattern is calibrated in advance corresponding to the pixel position in the pixel array; the memory stores the pixel position corresponding to the aiming pattern in advance; and the processor is used for converting the depth data acquired by the depth detection module into a depth point cloud and calculating the size of the target object according to the depth point cloud corresponding to the pixel position.

7. The dimensional measurement method according to claim 6, wherein: the module of aiming includes the display screen, the fixed position of display screen shows there is the aiming pattern, the display screen is used for showing the image of target object, through aiming the target object of pattern includes: the attitude of the sizing device is adjusted so that the aiming pattern overlaps the image of the target item on the display screen.

8. The dimensional measurement method according to claim 6, wherein: further comprising: the aiming pattern is generated by a sizing application.

9. The dimensional measurement method according to claim 6, wherein: the processor calculates the vertex position of the target object in real time, and generates the cuboid envelope of the target object on the display screen in real time based on the vertex position of the target object.

10. The dimensional measurement method according to claim 6, wherein: when a plurality of articles exist in the field range of the depth detection module, the processor cuts and fits the depth point cloud into a plurality of point cloud planes, and the articles corresponding to the point cloud planes overlapped with the aiming patterns are target articles.

Images