# US 2021/0279478 A1 · Vision system and method for a motor vehicle

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

- Number: US 2021/0279478 A1
- Document ID: US20210279478A1
- Status: Published application
- Kind: A1
- Application number: 16/328,583
- Filed: 2017-08-30
- Published: 2021-09-09
- Earliest claimed: 2016-08-31
- Assignees: Veoneer Sweden AB
- Applicants: Veoneer Sweden AB
- Inventors: ERIK NARBY; MARTIN NILSSON
- Web page: https://patentgrep.com/patent/US20210279478A1

## CPC

- G06V 20/588: {Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road}
- G06V 10/44: Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V 10/56: relating to colour
- G06V 20/56: exterior to a vehicle by using sensors mounted on the vehicle
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A vision system (10) for a motor vehicle having an imaging apparatus (11) to capture images from a surrounding of the motor vehicle and a data processing device (14) performing image processing on images captured by the imaging apparatus (11) and to control at least one driver assistance device (18) depending on a result of the image processing. The image processing having object detection through edge detection and the object detection having decomposing (21, 22) images captured by the imaging apparatus into a plurality of decomposed images (23, 24) having different color characteristics. Separate edge detection (35, 36) is performed on the decomposed images. The edges (37, 38) from the separate edge detections are merged (39) into a common list of edges. An edge distinguishability measure is calculated for each of the edges (37, 38), and the merging of the edges (37, 38) is based on the edge distinguishability measure.

## Claims

### Claim 1

A vision system for a motor vehicle, comprising an imaging apparatus adapted to capture images from a surrounding of the motor vehicle and a data processing device adapted to perform image processing on images captured by the imaging apparatus and to control at least one driver assistance device depending on a result of the image processing, wherein the image processing comprises object detection through edge detection, the object detection comprising decomposing images captured by the imaging apparatus into a plurality of decomposed images having different color characteristics, performing separate edge detection on the decomposed images, and merging detected edges from the separate edge detections into a common list of edges, an edge distinguishability measure is calculated for each of the edges, and the merging of the edges is based on the edge distinguishability measure.

### Claim 2 (depends on 1)

The vision system as claimed in claim 1 further comprising the merging of the edges is based on selecting the edge among the edges from different decomposed images providing a highest distinguishability measure.

### Claim 3 (depends on 1)

The vision system as claimed in claim 1 further comprising in that a gradient image is calculated on each of the decomposed images.

### Claim 4 (depends on 3)

The vision system as claimed in claim 3 further comprising in that the edge distinguishability measure is a gradient amplitude of each of the detected edges.

### Claim 5 (depends on 3)

The vision system as claimed in claim 3 further comprising a normalized gradient is calculated for each of the detected edges.

### Claim 6 (depends on 5)

The vision system as claimed in claim 5 further comprising the normalized gradient is calculated by dividing the gradient value of the detected edges by a corresponding edge threshold value.

### Claim 7 (depends on 1)

The vision system as claimed in claim 1 further comprising an edge threshold is calculated for each of the decomposed images.

### Claim 8 (depends on 7)

The vision system as claimed in claim 7 further comprising, the edge threshold is calculated from a statistical center value of gradient values, including a median of gradient values of other values.

### Claim 9 (depends on 7)

The vision system as claimed in claim 7 further comprising the edge detection is performed by thresholding a gradient image against the edge threshold.

### Claim 10 (depends on 1)

The vision system as claimed in claim 1 further comprising duplicates of the detected edges are removed from all of the edges detected in each of the decomposed images.

### Claim 11 (depends on 1)

The vision system as claimed in claim 1 further comprising two of the detected edges are considered duplicates if their positions are within a specified limit in world coordinates.

### Claim 12 (depends on 1)

The vision system as claimed in claim 1 further comprising the decomposed images comprise an illumination intensity image and a weighted color image.

### Claim 13 (depends on 12)

The vision system as claimed in claim 12 further comprising the weighted color image has the same color as the object in the environment of the motor vehicle to be detected.

### Claim 14 (depends on 12)

The vision system as claimed in claim 12 further comprising the weighted color image is a yellow image calculated from an RGB image by a linear combination between (R+G+2×(Bmax−B))/4 and (R+(Bmax−B))/2.

### Claim 15

A vision method for a motor vehicle, comprising capturing images from a surrounding of the motor vehicle, performing image processing on captured images and controlling at least one driver assistance device depending on a result of the image processing, wherein the image processing comprises object detection through edge detection, the object detection comprising decomposing captured images into a plurality of decomposed images having different color characteristics, performing separate edge detection on the decomposed images, and merging the edges from the separate edge detections into a common list of edges, an edge distinguishability measure is calculated for each of the edges, and the merging of the edges is based on the edge distinguishability measure.
