Volume measurement method and volume measurement apparatus

Abstract

The present disclosure provides a volume measurement method, which comprises the following steps. A transparent carrier is disposed above a background, and an object to be measured is placed on the transparent carrier so as to maintain a spacing distance between a top surface of the object to be measured and the background. An image of the object to be measured placed on the transparent carrier is captured by a depth camera disposed above the transparent carrier, wherein the image of the object to be measured comprises depth information, and the spacing distance is greater than or equal to a minimum depth measurement limit of the depth camera. The volume of the object to be measured is calculated according to the depth information in the image of the object to be measured. The present disclosure also provides a volume measurement device to perform the aforementioned volume measurement method.

Description

Technical Field

[0001] This disclosure relates to a volume measurement method and a volume measurement device, and more particularly to a volume measurement method and a volume measurement device capable of measuring small articles to be measured. Background Technology

[0002] With the rapid development of e-commerce platforms, the volume of goods transported in Business to Consumer (B2C) and Customer to Customer (C2C) consumption models is increasing year by year. For logistics and warehousing operators, the volume of goods directly affects the overall logistics and warehousing costs. Therefore, both warehousing and transportation rely on volumetric identification systems to monitor the storage and transportation status of goods. Depth camera technology is maturing, and with continuous improvements in software and hardware, there are now applications that can perform simple distance measurements using a mobile phone camera. Several volumetric measurement systems based on depth cameras are also available. Because image recognition does not require overly complex equipment and has few installation restrictions, the implementation costs for logistics operators can be significantly reduced. Existing volumetric measurement systems on the market require linear scanning technology to measure goods with a thickness of less than 5 cm and that are not regular rectangles. While linear scanning technology can obtain accurate volumetric measurement data, the equipment is expensive, requiring significant investment for widespread deployment by logistics operators. Summary of the Invention

[0003] This disclosure provides a method for measuring volume and a device for measuring volume.

[0004] This disclosure provides a volume measurement method, comprising the following steps: A transparent carrier is placed above a background, and an object to be measured is placed on the transparent carrier such that a gap is maintained between the top surface of the object and the background. An image of the object to be measured placed on the transparent carrier is captured by a depth camera positioned above the transparent carrier, wherein the image of the object to be measured includes depth information, and the gap is greater than or equal to the minimum depth measurement limit of the depth camera. The volume of the object to be measured is calculated based on the depth information in the image of the object.

[0005] In some embodiments of this disclosure, calculating the volume of the object under test based on depth information in an image of the object under test includes: identifying the sides of the object under test via a side detection calculation module in the processor; removing noise caused by the transparent carrier via a noise comparison module in the processor; and calculating the volume of the object under test via a volume calculation module in the processor based on the depth information and the sides of the object under test. In some embodiments of this disclosure, identifying the sides of the object under test via a side detection calculation module in the processor includes: binarizing and removing background noise from an image of the object under test via the side detection calculation module; removing noise from the binarized and background-removed image via the side detection calculation module; identifying the distribution range of the object under test in the image via the side detection calculation module; and performing surge detection on all pixels distributed within the distribution range of the object under test from the boundary of the distribution range towards the center of the distribution range via the side detection calculation module to identify the sides of the object under test in the image. In some embodiments of this disclosure, surge detection includes: analyzing the depth information of any two neighboring pixels from the boundary of the distribution range towards the center of the distribution range via a side detection calculation module, until all pixels located on the boundary of the distribution range have been analyzed. When the depth value of a pixel is 0, it is considered a surge; when the depth value difference between two neighboring pixels is 1.5 cm, it is considered the side of the object under test. In some embodiments of this disclosure, removing noise caused by the transparent carrier via a noise comparison module in the processor includes: capturing a first image of the background via a depth camera; capturing a second image of the transparent carrier positioned above the background via a depth camera; comparing the first image and the second image via a noise comparison module to obtain a background noise image; capturing an image of the object under test placed on the transparent carrier via a depth camera; and subtracting the first image and the background noise image from the image via the noise comparison module.

[0006] This disclosure provides a volume measurement device, comprising a transparent carrier, a depth camera, and a processor. The transparent carrier is disposed above a background and is adapted to hold an object to be measured. The depth camera is disposed above the transparent carrier, wherein the top surface of the object to be measured maintains a gap distance from the background, and the gap distance is greater than or equal to the minimum depth measurement limit of the depth camera. The processor calculates the volume of the object to be measured based on the image of the object.

[0007] In some embodiments of this disclosure, the depth camera includes an infrared depth camera, and the transparent carrier includes an infrared-permeable material. In some embodiments of this disclosure, the infrared-permeable material includes glass, acrylic, polymethyl methacrylate (PMMA), or a combination of the foregoing. In some embodiments of this disclosure, the distance maintained between the top surface of the object under test and the background is greater than or equal to 10 cm. In some embodiments of this disclosure, the processor includes a side detection calculation module, a noise comparison module, and a volumetric calculation module.

[0008] Based on the above, the embodiments of this disclosure can measure the volume of small test items without significant modifications to the volume measurement equipment. Compared to linear scanning technology, the volume measurement equipment of this disclosure has a considerable advantage in terms of setup cost. Furthermore, the volume measurement method and equipment in the embodiments of this disclosure can quickly measure the volume of small test items. Attached Figure Description

[0009] A thorough understanding of the various aspects of this disclosure is best achieved by reading the following detailed description in conjunction with the accompanying drawings. It should be noted that, in accordance with standard practice in the industry, the features are not drawn to scale. In fact, the dimensions of the features may be arbitrarily increased or decreased for clarity of explanation.

[0010] Figure 1 This is a cross-sectional schematic diagram of the volume measuring device according to an embodiment of the present disclosure.

[0011] Figures 2 to 10 This is the side detection calculation process according to the embodiments of this disclosure.

[0012] Figures 11 to 12 This is the noise comparison process according to the embodiments of this disclosure. Detailed Implementation

[0013] The following disclosure provides numerous different embodiments or examples to implement various features of the provided subject matter. Specific examples of components and arrangements are set forth below to simplify this disclosure. Of course, these are merely examples and are not intended to be limiting. For example, forming a first feature on or over a second feature in the following description may include embodiments in which the first and second features are formed in direct contact, and may also include embodiments in which an additional feature may be formed between the first and second features such that the first and second features are not in direct contact. Furthermore, reference numerals and / or letters may be repeated throughout this disclosure. Such repetition is for the purpose of brevity and clarity and is not, in itself, indication of a relationship between the various embodiments and / or configurations discussed.

[0014] Figure 1 This is a cross-sectional schematic diagram of the volume measuring device according to an embodiment of the present disclosure.

[0015] Please refer to Figure 1 The volume measurement device 100 of this embodiment includes a transparent carrier 110, a depth camera 120, and a processor 130. The transparent carrier 110 is disposed above a background 140 and is adapted to support the object to be measured 150. The depth camera 120 is disposed above the transparent carrier 110, wherein the top surface 150a of the object to be measured 150 and the background 140 maintain a distance D1, and the distance D1 is greater than or equal to the minimum depth measurement limit of the depth camera 120. In other words, the supporting surface 110a of the transparent carrier 110 and the background 140 maintain a distance D2, and the sum of the distance D2 and the thickness of the object to be measured 150 is greater than or equal to the minimum depth measurement limit of the depth camera 120. Furthermore, the processor 130 calculates the volume of the object to be measured 150 based on the image of the object to be measured 150 captured by the depth camera 120.

[0016] In some embodiments of this disclosure, the transparent carrier 110 includes a transparent substrate and an infrared-permeable material, such as glass, acrylic, polymethyl methacrylate (PMMA), or a combination of the foregoing that allows infrared light of a specific wavelength (e.g., 850 nm) to pass through. The depth camera 120 includes an infrared depth camera. For example, the depth camera includes a camera, an infrared emitter, and an infrared lens. In other embodiments, the depth camera 120 includes multiple cameras. By comparing the images recorded by the lenses of the multiple cameras using an algorithm, depth information for each pixel can be obtained, thereby establishing three-dimensional spatial information (i.e., position information and depth information). In other feasible embodiments, the depth camera 120 may further include an infrared projection device that projects infrared light onto the subject. The distance between the subject and the camera (i.e., depth information) can be calculated from the infrared light reflected from the subject. As described above, any camera capable of simultaneously capturing position information and depth information can be used as the depth camera 120 in this embodiment.

[0017] In some embodiments of this disclosure, the processor 130 includes a side detection calculation module 132, a noise comparison module 134, and a volume calculation module 136. For example... Figure 1As shown, processor 130 is electrically connected to depth camera 120, and processor 130 is used to receive and process image signals from depth camera 120. In the image captured by depth camera 120, the three-dimensional spatial information of each pixel includes the pixel's position information (i.e., x-axis and y-axis coordinates) and the pixel's depth information (i.e., z-axis coordinate). The side detection calculation module 132 in processor 130 can receive and process the image captured by depth camera 120 to identify the side of the object under test 150. The noise comparison module 134 in processor 130 can remove noise caused by transparent carrier 110. Based on the depth information and side of the object under test 150, the volume calculation module 136 in processor 130 can calculate the volume of the object under test 150.

[0018] In some embodiments of this disclosure, the background 140 located below the transparent carrier 110 includes the ground, a carrier plate, a background cloth, etc.

[0019] In some embodiments of this disclosure, the thickness of the object to be measured 150 may be less than, equal to, or greater than the minimum depth measurement limit of the depth camera 120. In other words, the thickness of the object to be measured 150 is not limited by the minimum depth measurement limit of the depth camera 120. Here, the minimum depth measurement limit of the depth camera 120 refers to the minimum depth that the depth camera 120 can detect. In some embodiments of this disclosure, the minimum depth measurement limit of the depth camera 120 is approximately 10 cm, while the provision of the transparent carrier 110 can exceed the minimum depth measurement limit of the depth camera 120, enabling the volumetric measuring device 100 to measure the thickness of the object to be measured 150, particularly when the thickness of the object to be measured 150 is less than the minimum depth measurement limit of the depth camera 120 (e.g., 10 cm).

[0020] In some embodiments, the volume measurement device 100 further includes a support 160 for maintaining a distance D1 between the top surface 150a of the item to be measured 150 and the background 140. In some embodiments of this disclosure, the distance D1 maintained between the top surface 150a of the item to be measured 150 and the background 140 is greater than or equal to 10 cm.

[0021] Figures 2 to 10 This is the side detection calculation process according to the embodiments of this disclosure.

[0022] Please refer to Figure 1 and Figure 2First, a test object 150 of unknown thickness is placed on a transparent carrier 110, maintaining a distance D1 between the top surface 150a of the test object 150 and the background 140. Next, a depth camera 120 positioned above the transparent carrier 110 and the test object 150 captures an image (IMG-1) of the test object 150 placed on the transparent carrier 110. Figure 2 (As shown). In some embodiments of this disclosure, the depth camera 120 is positioned directly above the object under test 150. In other embodiments of this disclosure, the depth camera 120 can be configured according to actual needs, as long as it can completely capture the image IMG-1 of the object under test 150. In other words, the depth camera 120 does not need to be positioned directly above the object under test 150. In some embodiments of this disclosure, the image IMG-1 of the object under test 150 captured by the depth camera 120 includes depth information embedded in the two-dimensional image IMG-1. That is, the image IMG-1 of the object under test 150 captured by the depth camera 120 includes three-dimensional spatial information, and the aforementioned three-dimensional spatial information includes the position information (i.e., x-axis coordinates and y-axis coordinates) of each pixel in the image IMG-1 and the depth information (i.e., z-axis coordinates) of each pixel in the image IMG-1. Here, the depth information represents the distance between each pixel and the depth camera 120 in the z-axis direction.

[0023] from Figure 2 It is known that when the depth camera 120 captures the image IMG-1 of the object under test 150, irregular black noise (e.g., irregular black lines) will appear on the side of the object under test 150 in the image IMG-1 due to factors such as the mounting position of the depth camera 120, the relative position between the depth camera 120 and the object under test 150, the shooting angle, and the lighting. This irregular black noise will affect the side detection of the object under test 150.

[0024] After the depth camera 120 captures the image IMG-1 of the object under test 150, the side detection calculation module 132 in the processor 130 will confirm the side of the object under test. The following will be accompanied by... Figures 3 to 10 The calculation process for side detection of the test item 150 is explained in detail. The side detection calculation process for the test item 150 includes: image binarization and background removal processing (such as...). Figure 3 (as shown), image denoising processing (such as) Figure 4 As shown), confirm the distribution range of the test item 150 (e.g. Figure 5 (as shown) and surge detection (such as) Figures 6 to 7 (as shown in the figure) and other steps.

[0025] Please refer to Figure 1 , Figure 2and Figure 3 The side detection calculation module 132 in the processor 130 processes the IMG-1 image of the object to be measured 150 captured by the depth camera 120 (e.g., ...). Figure 2 (As shown) Binarization and background removal are performed to obtain the result as shown. Figure 3 The image shown is IMG-2.

[0026] Please refer to Figure 1 , Figure 3 and Figure 4 The side detection calculation module 132 in the processor 130 processes the binarized and background-removed image IMG-2 (such as... Figure 3 (As shown) Perform noise reduction processing to obtain as Figure 4 The image IMG-3 is shown. In some embodiments of this disclosure, the aforementioned denoising process includes erosion and dilation image processing techniques to remove jagged edges in the image IMG-2.

[0027] Please refer to Figure 1 and Figure 5 The side detection and calculation module 132 in the processor 130 confirms the distribution range R of the object under test 150 in the image IMG-3 (as shown in the dashed box). Here, the confirmation step of the distribution range of the object under test 150 is mainly used to confirm that the location of the object under test 150 is within the distribution range R, rather than outside the distribution range R.

[0028] Please refer to Figure 1 as well as Figures 6 to 7 The side detection calculation module 132 in the processor 130 performs surge detection on all pixels distributed within the distribution range R in the image IMG-3. In some embodiments of this disclosure, the side detection calculation module 132 in the processor 130 performs surge detection on all pixels distributed within the distribution range R from the boundary of the distribution range R towards the center of the distribution range (e.g., ...). Figure 6 (As indicated by the arrow within the distribution range R), to confirm the true side E of the object under test 150 in the image IMG-3 (e.g., Figure 7 (As shown). The following will be paired with... Figures 8 to 10 A detailed explanation of the calculation process for surge detection is provided.

[0029] Please refer to Figure 1 and Figure 8 The side detection and calculation module 132 in processor 130 obtains the point array information Box of all contour pixels distributed on the boundary of the distribution range R. Here, the point array information Box of all contour pixels is defined as:

[0030] Box=[P1, P2, P3...Pn],

[0031] The coordinates of each pixel Pi are (xi, yi), and i = 1, 2, 3, ..., n;

[0032] The checkpoint Pcheck is initially set to pixel Pi, meaning Pcheck = (xcheck, ycheck) = (xi, yi); and

[0033] The coordinates of the center point Pcenter of the distribution range R are (xcenter, ycenter).

[0034] Please refer to Figure 1 and Figure 9 The next pixel Pnext is calculated by the side detection calculation module 132 in the processor 130. The next pixel Pnext is calculated from the relative relationship between the check point Pcheck and the center point Pcenter. The calculation method of the next pixel Pnext is as follows:

[0035] Pnext = (xnext, ynext)

[0036] Pcheck=(xcheck,ycheck)=(xi,yi)

[0037] Pcenter = (xcenter, ycenter)

[0038] if (xcenter > xcheck)

[0039] xnext==xcheck+k

[0040] else if (xcenter < xcheck)

[0041] xnext == xcheck-k

[0042] if (ycenter > ycheck)

[0043] ynext = ycheck + k

[0044] else if (ycenter < ycheck)

[0045] ynext = ycheck-k

[0046] Where k is the inspection interval.

[0047] Please refer to Figure 1 and Figure 10The side detection calculation module 132 in the processor 130 checks whether the depth value of pixel Pnext is 0. If the depth value Znext of pixel Pnext is 0, then the checkpoint Pcheck is regarded as a surge. Then, the checkpoint Pcheck is updated to Pnext (that is, Pcheck = Pnext), and the process returns to... Figure 9 The steps disclosed herein are as follows: If the depth value Znext of pixel Pnext is not 0, check whether the difference in depth value between pixel Pnext and checkpoint Pcheck exceeds the side height increase threshold Zmax (e.g., 1.5 cm). If the difference in depth value between pixel Pnext and checkpoint Pcheck exceeds the side height increase threshold Zmax, then pixel Pnext is considered part of the side E. Next, checkpoint Pcheck is updated to Pnext (i.e., Pcheck = Pnext), and the process returns to... Figure 9 The steps disclosed herein; conversely, when the depth difference between pixel Pnext and checkpoint Pcheck does not exceed the side height increase threshold Zmax, and is not a surge, then pixel Pnext is considered a valid point, and then pixel Pi is replaced with Pnext (i.e., let Pi = Pnext). As mentioned above, during the calculation process, if checkpoint Pcheck exceeds center point Pcenter, it means that no better solution can be found. In this case, the original coordinates of pixel Pi are retained.

[0048] from Figures 8 to 10 It is clear that the side detection calculation module 132 in the processor 130 can analyze the depth information of any two neighboring pixels Pcheck and Pnext from the boundary of the distribution range R towards the center of the distribution range R, until all pixels Pi within the distribution range R have been analyzed. In some embodiments of this disclosure, the coordinates of each pixel P1, P2, P3, ..., Pn in the point array information Box of the contour pixels can be obtained respectively through... Figures 8 to 10 The algorithm described herein is updated or retained to obtain the effective side E of the test item 150 (drawn in...). Figure 7 ).

[0049] Figures 11 to 12 This is the noise comparison process according to the embodiments of this disclosure.

[0050] Please refer to Figure 1 , Figure 11 and Figure 12The noise comparison module 134 in the processor 130 can be used to remove noise caused by the transparent carrier 110. The noise comparison module 134 performs the following steps to remove noise caused by the transparent carrier 110: First, a first image IMG-A of the background 140 is captured via the depth camera 120; then, a second image IMG-B of the transparent carrier 110 positioned above the background 140 is captured via the depth camera 120; subsequently, the noise comparison module 134 compares the first image IMG-A and the second image IMG-B to obtain a background noise image IMG-C, as shown below. Figure 11 As shown. After obtaining the background noise image IMG-C, the depth camera 120 then captures the image IMG-D of the test object 150 placed on the transparent carrier 110. Then, the noise comparison module 134 subtracts the first image IMG-A and the background noise image IMG-C from the image IMG-D to obtain the denoised image IMG-E.

[0051] The higher the light transmittance of the transparent carrier 110, the less noise it has. If the transparent carrier 110 has scratches or strong light reflection, it will increase the noise in the image IMG-B. In this embodiment, by cross-comparing the background before and after the transparent carrier 110 is set up, the noise caused by the transparent carrier 110 can be removed, thereby improving the accuracy of the volume measurement of the item under test 150.

[0052] After removing the noise caused by the transparent carrier 110, based on the depth information in the IMG-E image of the object under test 150 and the effective side edge E (plotted on...), Figure 7 The volume of the object to be tested 150 is calculated by the volume calculation module 136 in the processor 130.

[0053] Based on the above, the embodiments of this disclosure can measure the volume of small test items without significant modifications to the volume measurement equipment. Compared to linear scanning technology, the volume measurement equipment of this disclosure has a considerable advantage in terms of setup cost. Furthermore, the volume measurement method and equipment in the embodiments of this disclosure can quickly measure the volume of small test items.

[0054] The foregoing outlines features of several embodiments to enable those skilled in the art to better understand various aspects of this disclosure. Those skilled in the art will understand that they can readily use this disclosure as a basis for designing or modifying other processes and structures to achieve the same purposes and / or realize the same advantages as the embodiments described herein. Those skilled in the art should also recognize that these equivalent constructions do not depart from the spirit and scope of this disclosure, and that various changes, substitutions, and modifications can be made herein without departing from the spirit and scope of this disclosure.

Claims

1. A method for measuring volume, characterized in that, include: A transparent carrier is placed above a background, and the item to be tested is placed on the transparent carrier so that the top surface of the item to be tested is kept at a distance from the background, wherein the distance between the top surface of the item to be tested and the background is maintained by a support. An image of the object under test placed on the transparent carrier is captured by a depth camera positioned above the transparent carrier, wherein the image of the object under test includes depth information, and the interval distance is greater than or equal to the minimum depth measurement limit of the depth camera. as well as The volume of the test item is calculated based on the depth information in the image of the test item.

2. The volume measurement method according to claim 1, characterized in that, Calculating the volume of the test item based on the depth information in the image of the test item includes: The side of the object under test is confirmed by the side detection calculation module in the processor; The noise caused by the transparent carrier is removed via a noise comparison module in the processor; as well as The volume of the test object is calculated by the volume calculation module in the processor based on the depth information and the side of the test object.

3. The volume measurement method according to claim 2, characterized in that, The side detection calculation module in the processor confirms that the side of the object under test includes: The image of the object under test is binarized and background removed by the side detection calculation module. The side detection calculation module performs noise reduction processing on the image after binarization and background removal; The distribution range of the object under test is determined in the image via the side detection calculation module; and The side detection calculation module performs surge detection on all pixels distributed within the distribution range of the object under test from the boundary of the distribution range to the center of the distribution range to identify the side of the object under test in the image.

4. The volume measurement method according to claim 3, characterized in that, The surge detection includes: The side detection calculation module analyzes the depth information of any two neighboring pixels from the boundary of the distribution range toward the center of the distribution range until all pixels located on the boundary of the distribution range have been analyzed. When the depth value of a pixel is 0, it is considered a surge. When the depth difference between two adjacent pixels is 1.5 cm, it is considered as the side of the object under test.

5. The volume measurement method according to claim 2, characterized in that, Removing noise caused by the transparent carrier via the noise comparison module in the processor includes: A first image of the background is captured by the depth camera; A second image of the transparent carrier positioned above the background is captured by the depth camera; The first image and the second image are compared via the noise comparison module to obtain a background noise image; The depth camera captures an image of the object under test placed on the transparent carrier; and The first image and the background noise image are subtracted from the image via the noise comparison module.

6. A volume measuring device, characterized in that, include: A transparent carrier is placed above the background, and the transparent carrier is suitable for holding the item to be tested. A support, wherein the distance between the top surface of the object to be tested and the background is maintained by the support; A depth camera is positioned above the transparent carrier, wherein the top surface of the object to be measured is maintained at a distance from the background, and the distance is greater than or equal to the minimum depth measurement limit of the depth camera. as well as The processor calculates the volume of the object under test based on the image of the object under test captured by the depth camera.

7. The volume measuring device according to claim 6, characterized in that, The depth camera includes an infrared depth camera, and the transparent carrier includes an infrared-permeable material.

8. The volume measuring device according to claim 7, characterized in that, The infrared-transmittable materials include glass, acrylic, polymethyl methacrylate, or a combination of the foregoing.

9. The volume measuring device according to claim 6, characterized in that, The distance maintained between the top surface of the object to be tested and the background is greater than or equal to 10 centimeters.

10. The volume measuring device according to claim 6, characterized in that, The processor includes a side detection calculation module, a noise comparison module, and a volume calculation module.

Images