# US 2019/0378248 A1 · Distance measuring apparatus, vibration measuring apparatus, and industrial computed tomography apparatus

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0378248 A1
- Document ID: US20190378248A1
- Status: Published application
- Kind: A1
- Application number: 16/285,997
- Filed: 2019-02-26
- Published: 2019-12-12
- Earliest claimed: 2018-06-07
- Assignees: Kabushiki Kaisha Toshiba
- Applicants: Kabushiki Kaisha Toshiba
- Inventors: Takashi IDA; Kenzo ISOGAWA; Atsushi MATSUMURA
- Web page: https://patentgrep.com/patent/US20190378248A1

## CPC

- G06T 5/70: Denoising; Smoothing
- G01S 7/4808: {Evaluating distance, position or velocity data}
- G01S 17/42: Simultaneous measurement of distance and other co-ordinates (indirect measurement G01S17/46)
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/048: Activation functions
- G06N 3/09: Supervised learning
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

According to one embodiment, a distance measuring apparatus includes a processing circuit and a memory. The processing circuit generates a distance image by measuring distances to an obstacle by using the time until the emitted light is reflected from the obstacle and returned as reflected light. The processing circuit generates an intensity image by measuring the intensity of the reflected light or the intensity of the environmental light. The memory stores a learned model for generating a denoise image, in which noise of the distance image is reduced, based on the distance image and the intensity image. The processing circuit inputs the distance image and the intensity image to the learned model, and generates the

## Claims

### Claim 1

A distance measuring apparatus comprising:
a processing circuit which generates a distance image by measuring distances to an obstacle by using time until emitted light is reflected from the obstacle and returned as reflected light, and generates an intensity image by measuring intensity of the reflected light or intensity of environmental light; and
a memory which stores a learned model for generating a denoise image, in which noise of the distance image is reduced, based on the distance image and the intensity image,
wherein the processing circuit inputs the distance image and the intensity image to the learned model and generates the denoise image by using the learned model.

### Claim 2 (depends on 1)

The distance measuring apparatus according to claim 1, wherein the learned model is a neural network, and
the processing circuit inputs the distance image and the intensity image to different channels of an input layer in the neural network.

### Claim 3 (depends on 1)

The distance measuring apparatus according to claim 1, wherein the learned model is a neural network, and
the processing circuit inputs the distance image to an input layer in the neural network and inputs the intensity image to at least one of a plurality of intermediate layers in the neural network.

### Claim 4 (depends on 1)

The distance measuring apparatus according to claim 1, wherein the intensity image has a pixel value correlated with a noise amount or a blur amount with respect to each of the distances corresponding to a plurality of pixels, respectively.

### Claim 5

A vibration measuring apparatus comprising:
an acceleration sensor which measures a vibration along time series;
a memory which stores a learned model for generating denoise data, in which noise of vibration is reduced for each frequency, based on noise correlation data correlated with the noise of the vibration for the each frequency of the vibration and data for the each frequency of the vibration; and
a processing circuit which inputs the vibration data and the noise correlation data to the learned model and generates the denoise data by using the learned model.

### Claim 6 (depends on 5)

The vibration measuring apparatus according to claim 5, wherein the memory further stores a second learned model for generating the noise correlation data based on the vibration data, and
the processing circuit inputs the vibration data to the second learned model and generates the noise correlation data by using the second learned model,

### Claim 7

An industrial computed tomography apparatus comprising:
a memory which stores a learned model functioned to generate a denoise image, in which noise of a structure image or noise of a sinogram is reduced, based on the structure image reconstructed based on projection data collected with respect to the structure or the sinogram generated based on the projection data, and a noise correlation map correlated with noise included in the structure image or the sinogram; and
a processing circuit which inputs the noise correlation map and the structure image or the sinogram to the learned model, and generates the denoise image, in which noise of the structure image or the noise of the sinogram is reduced, by using the learned model.

### Claim 8 (depends on 7)

The industrial computed tomography apparatus according to claim 7, wherein the noise correlation map is a transmission length map indicating a transmission length of the structure along a ray of X-rays passing through the structure for each rotation angle of a rotary table, on which the structure is placed, in the structure image, or a transmission length reconstruction image reconstructed by using the transmission length map.