Information processing device, and information processing method

Abstract

To reduce influence due to noise about a single detection point group data.SOLUTION: According to an embodiment, an information processing device includes a division part for dividing a two-dimensional area corresponding to two-dimensional detection point group data being a set of detection points detected as the position of an object into a plurality of small areas of a prescribed size, a measurement part for measuring the number of detection points in each small area, a detection part for using the number of detection points to detect an edge in the unit of the small area about the detection point group data, and a selection part for selecting the detection points included in the small area detected as the edge.SELECTED DRAWING: Figure 11

Description

[Technical Field]

[0001] The present embodiment relates to an information processing device and an information processing method. [Background technology]

[0002] Conventionally, there has been a technology for detecting an object based on data obtained from a predetermined sensor (e.g., a millimeter wave sensor). Specifically, for example, first, a transmission wave is transmitted from the sensor into a predetermined space, and reflected waves generated by reflection from an object present in the predetermined space are received, and three-dimensional detection point cloud data is calculated based on the reception results. Then, an object is detected based on the three-dimensional detection point cloud data. Alternatively, for example, the three-dimensional detection point cloud data may be divided into predetermined widths in a predetermined dimensional direction and treated as multiple pseudo-two-dimensional detection point cloud data, and an object may be detected based on the multiple two-dimensional detection point cloud data.

[0003] However, such detection point cloud data may contain noise, so it is important to reduce the influence of noise. For this purpose, there is a conventional technique for reducing the influence of noise based on the state of temporal changes in the detection point cloud. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-196195 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the above-mentioned conventional technology requires information on the temporal changes of the detection point cloud. In other words, multiple pieces of detection point cloud data are required in time series. Therefore, it is not possible to reduce the influence of noise on a single piece of detection point cloud data.

[0006] Therefore, one of the objectives of this embodiment is to provide an information processing device and an information processing method that can reduce the influence of noise on a single piece of detection point cloud data. [Means for solving the problem]

[0007] The information processing device of this embodiment includes a division unit that divides a two-dimensional area corresponding to two-dimensional detection point cloud data, which is a collection of detection points detected as the positions of objects, into a plurality of small areas of a predetermined size; a measurement unit that measures the number of detection points for each small area; a detection unit that detects edges for the detection point cloud data using the number of detection points for each small area; and a selection unit that selects the detection points included in the small areas detected as the edges.

[0008] According to this configuration, for a single detection point cloud data, it is possible to extract (select) with high accuracy the detection points contained in a small area detected as an edge, i.e., the contour part of an object, thereby reducing the influence of noise. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram illustrating a functional configuration of an information processing apparatus according to an embodiment. [Figure 2] FIG. 2 is an explanatory diagram of an outline of processing in the embodiment. [Figure 3] FIG. 3 is a first explanatory diagram of the details of the processing in the embodiment. [Figure 4] FIG. 4 is a second explanatory diagram of the details of the processing in the embodiment. [Figure 5] FIG. 5 is a third explanatory diagram of the details of the processing in the embodiment. [Figure 6] FIG. 6 is a fourth explanatory diagram of the details of the processing in the embodiment. [Figure 7] FIG. 7 is a fifth explanatory diagram of the details of the processing in the embodiment. [Figure 8] FIG. 8 is a sixth explanatory diagram of the details of the processing in the embodiment. [Figure 9] FIG. 9 is a seventh explanatory diagram illustrating details of the processing in the embodiment. [Figure 10] FIG. 10 is an eighth explanatory diagram illustrating details of the processing in the embodiment. [Figure 11] FIG. 11 is a flowchart illustrating processing by the information processing apparatus according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an information processing device and an information processing method according to embodiments of the present invention will be described with reference to the drawings. Fig. 1 is a block diagram showing the functional configuration of an information processing device 1 according to an embodiment. Fig. 2 is an explanatory diagram showing an overview of processing in the embodiment. Figs. 3 to 10 are explanatory diagrams showing details of processing in the embodiment.

[0011] 1 is a block diagram showing the functional configuration of an information processing device 1 according to an embodiment. The information processing device 1 is a computer device, and includes a processing unit 2, a storage unit 3, an input unit 4, a display unit 5, and a communication unit 6.

[0012] Here, an overview of the processing executed by the information processing device 1 will be described with reference to Fig. 2. From three-dimensional detection point cloud data G, which is a collection of detection points P detected as the positions of an object using a predetermined sensor (for example, a millimeter wave sensor) as shown in Fig. 2(a), detection points P corresponding to the contour portion of the object, as shown in Fig. 2(b), are extracted (selected). Details of the processing will be described below.

[0013] Returning to FIG. 1, the storage unit 3 is realized by, for example, a random access memory (RAM), a read only memory (ROM), a solid state drive (SSD), a hard disk drive (HDD), etc., and stores various types of information.

[0014] The storage unit 3 stores, for example, the operation program, various data, and various calculation results of the processing unit 2. The storage unit 3 also stores three-dimensional detection point cloud data calculated based on the reception results of reflected waves generated when a transmission wave transmitted into a predetermined space by a predetermined sensor (for example, a millimeter wave sensor) is reflected by an object present in the predetermined space. The storage unit 3 also stores predetermined widths for dividing a three-dimensional region corresponding to the three-dimensional detection point cloud data in each of the three-dimensional directions.

[0015] The processing unit 2 is realized by, for example, a CPU (Central Processing Unit) and executes various information processes. The processing unit 2 includes, for example, an acquisition unit 21, a division unit 22, a measurement unit 23, a detection unit 24, a selection unit 25, and a control unit 26 as functional components.

[0016] The acquisition unit 21 acquires various types of information. For example, the acquisition unit 21 acquires three-dimensional detection point cloud data (FIG. 3) from the storage unit 3.

[0017] As shown in Fig. 4, the dividing unit 22 divides a three-dimensional region corresponding to the three-dimensional detection point cloud data into regions with a predetermined width W in each three-dimensional direction. In the example of Fig. 4, the three-dimensional region is divided into eight regions in each dimension.

[0018] In the following processing, for the three-dimensional detection point cloud data, the detection point cloud data for each layer in a predetermined direction (the Z dimension in the examples from FIG. 5 onward) is treated as two-dimensional detection point cloud data corresponding to a two-dimensional area divided into a plurality of small areas of a predetermined size. For example, in FIG. 5, (a) to (d) are the two-dimensional detection point cloud data for the first to fourth layers in the Z direction, respectively. In other words, the dividing unit 22 divides the two-dimensional area corresponding to the two-dimensional detection point cloud data into a plurality of small areas of a predetermined size.

[0019] The measurement unit 23 measures the number of detection points for each small region for each piece of two-dimensional detection point cloud data. For example, in the example of Fig. 6, the result of measuring the number of detection points for each small region for the two-dimensional detection point cloud data shown in (a) is as shown in (b).

[0020] The detection unit 24 detects edges for each small region of the two-dimensional detection point cloud data using the number of detection points. In this case, for example, if each small region is regarded as a pixel and the number of detection points for each small region is regarded as the brightness of each pixel, the two-dimensional detection point cloud data can be treated as two-dimensional image data in the vertical and horizontal axis directions. Therefore, edges can be detected by applying a known edge detection method (e.g., the Canny algorithm) to two-dimensional image data. For example, in the example of FIG. 7, the result of edge detection for the two-dimensional detection point cloud data shown in (a) is as shown in (b). In FIG. 7(b), the symbol E indicates a small region detected as an edge.

[0021] The selection unit 25 selects detection points included in small regions detected as edges for each of the two-dimensional detection point cloud data. For example, in the example of FIG. 8, as shown in (a), detection points included in small regions that are not detected as edges are discarded. (b) is a diagram showing the number of detection points for each small region. Then, such processing by the selection unit 25 is performed for each layer in the Z direction as shown in FIG.

[0022] The control unit 26 executes various controls. For example, the control unit 26 displays the detection points selected by the selection unit 25 on the display unit 5. Fig. 10 is an example of such a display. With respect to the three-dimensional detection point cloud data G, the detection points P included in small regions detected as edges, that is, the outline portions of the object, are displayed.

[0023] Returning to FIG. 1, the input unit 4 is a means for the user to input information, and is realized by, for example, a keyboard or a mouse.

[0024] The display unit 5 is a means for displaying information, and is realized by, for example, an LCD (Liquid Crystal Display).

[0025] The communication unit 6 is a communication interface for communicating with an external device (not shown).

[0026] Next, the processing by the information processing device 1 will be described with reference to Fig. 11. Fig. 11 is a flowchart showing the processing by the information processing device 1 according to the embodiment.

[0027] In step S1, the acquisition unit 21 acquires three-dimensional detection point cloud data (FIG. 3) from the storage unit 3.

[0028] Next, in step S2, the dividing unit 22 divides the three-dimensional region corresponding to the three-dimensional detection point cloud data into regions with a predetermined width W in each of the three-dimensional directions (FIG. 4).

[0029] Next, in step S3, the division unit 22 associates each small region with each detection point (FIG. 5).

[0030] Next, in step S4, the measurement unit 23 counts the number of detection points for each small region for each of the two-dimensional detection point cloud data (FIG. 6).

[0031] Next, in step S5, the detection unit 24 detects edges for each of the two-dimensional detection point cloud data in units of small regions using the number of detection points (FIG. 7).

[0032] Next, in step S6, the selection unit 25 selects detection points included in small regions detected as edges for each of the two-dimensional detection point cloud data (FIGS. 8 and 9).

[0033] Next, in step S7, the control unit 26 displays the detection points selected in step S6 on the display unit 5 (FIG. 10).

[0034] As described above, according to the information processing device 1 of this embodiment, it is possible to extract (select) with high accuracy detection points contained in a small area detected as an edge, that is, the contour part of an object, from a single two-dimensional detection point cloud data, thereby reducing the influence of noise.

[0035] Specifically, for a single three-dimensional detection point cloud data, the detection point cloud data for each layer in a predetermined direction (the Z dimension in the examples of Figures 5 and subsequent figures) can be treated as two-dimensional detection point cloud data corresponding to a two-dimensional area divided into multiple small areas, thereby enabling subsequent processing to be performed.

[0036] For example, by displaying the detection points P contained in the small area detected as an edge as shown in Figure 10, it is easier for the viewer to grasp the shape (contour) of the object compared to when all detection points P are displayed as shown in Figure 3.

[0037] Furthermore, by adjusting the size of the division width of the three-dimensional detection point cloud data, the processing speed and the accuracy of the calculation results can be adjusted.

[0038] Furthermore, by treating the two-dimensional detection point cloud data as two-dimensional image data as described above, it is possible to detect edges by applying various known edge detection methods to two-dimensional image data.

[0039] The program executed by the information processing device 1 of this embodiment can be provided by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD (Compact Disc)-ROM (Read Only Memory), a flexible disk (FD), a CD-R (Recordable), or a DVD (Digital Versatile Disk).The program may also be provided or distributed via a network such as the Internet.

[0040] Although an embodiment of the present invention has been described above, this embodiment is presented as an example and is not intended to limit the scope of the invention. This novel embodiment can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. This embodiment and its modifications are included within the scope and spirit of the invention, and are also included in the invention and its equivalents as defined in the claims.

[0041] For example, the sensor for obtaining three-dimensional detection point cloud data is not limited to a millimeter wave sensor, and may be another type of sensor such as a LiDAR (Light Detection and Ranging) or an ultrasonic sensor.

[0042] Furthermore, when dividing a three-dimensional region corresponding to three-dimensional detection point cloud data in each three-dimensional direction, the size of the division width in each direction may be the same or different.

[0043] Furthermore, when dividing a three-dimensional region as described above, interval arithmetic processing is required. For interval arithmetic processing, various publicly available open source software (OSS) can be used, which reduces the programming effort.

[0044] Furthermore, for example, the accuracy of object detection can be further improved by linking the detected point cloud data relating to the same area inside the vehicle cabin with image data captured by an in-vehicle camera and performing image processing.

[0045] In the above embodiment, the case where two-dimensional detection point cloud data for each layer in the Z direction among the X, Y, and Z directions is processed has been described, but the present invention is not limited to this. Alternatively, two-dimensional detection point cloud data for each layer in the X direction or Y direction may be processed. Furthermore, two-dimensional detection point cloud data for each layer in a plurality of directions may be processed, and each processing result may be displayed.

[0046] The above series of processes may also be performed on a plurality of pieces of detection point cloud data in time series. [Explanation of symbols]

[0047] 1...information processing device, 2...processing unit, 3...storage unit, 4...input unit, 5...display unit, 6...communication unit, 21...acquisition unit, 22...division unit, 23...measurement unit, 24...detection unit, 25...selection unit, 26...control unit

Claims

1. a dividing unit that divides a two-dimensional area corresponding to two-dimensional detection point cloud data, which is a collection of detection points detected as positions of an object, into a plurality of small areas of a predetermined size; a measurement unit that counts the number of detection points for each small region; a detection unit that detects edges in the detection point cloud data by using the number of detection points in units of the small regions; a selection unit that selects the detection points included in the small regions detected as the edges; An information processing device comprising:

2. An acquisition unit that acquires three-dimensional detection point cloud data calculated based on a reception result of a reflected wave generated when a transmission wave transmitted to a predetermined space is reflected by an object present in the predetermined space, the dividing unit divides a three-dimensional region corresponding to the three-dimensional detection point cloud data into predetermined widths in each of the three-dimensional directions; In the subsequent processing, with respect to the three-dimensional detection point cloud data, the detection point cloud data for each layer in a predetermined one-dimensional direction is treated as the two-dimensional detection point cloud data corresponding to the two-dimensional area divided into the plurality of small areas of the predetermined size, the measurement unit measures the number of detection points for each small region for each of the two-dimensional detection point cloud data; the detection unit detects edges for each of the two-dimensional detection point cloud data in units of small regions using the number of detection points; The information processing apparatus according to claim 1 , wherein the selection unit selects, for each of the two-dimensional detection point cloud data, the detection points included in the small regions detected as the edges.

3. a division step in which a division unit divides a two-dimensional area corresponding to two-dimensional detection point cloud data, which is a collection of detection points detected as positions of an object, into a plurality of small areas of a predetermined size; a measuring step in which a measuring unit measures the number of the detection points for each small region; a detection step in which a detection unit detects edges in the detection point cloud data by using the number of detection points in units of the small regions; a selection step in which a selection unit selects the detection points included in the small regions detected as the edges; An information processing method including:

Images