# US 2013/0279749 A1 · System and method for detecting target rectangles in an image

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

- Number: US 2013/0279749 A1
- Document ID: US20130279749A1
- Status: Published application
- Kind: A1
- Application number: 13/860,362
- Filed: 2013-04-10
- Published: 2013-10-24
- Earliest claimed: 2012-04-10
- Inventors: Victor KAISER-PENDERGRAST
- Web page: https://patentgrep.com/patent/US20130279749A1

## CPC

- 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
- Y10S 901/46: Sensing device
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- Y10S: TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS

## Abstract

A system and method for detecting target rectangles in an image acquired from a camera or computer file in which average pixel values from small horizontal and vertical rectangles called sub-exlons are computed from an edge-detected and binarized version of the image. The sub-exlons are then used to scan for exlons, which are combinations of horizontal and vertical sub-exlons that make up the four possible types of corner (upper left, upper right, lower left, and lower right). Possible corners that are located are then matched up and a rectangle is confirmed by scanning for properly oriented horizontal and vertical connecting lines composed from the previously computed sub-exlons. All confirmed rectangles are stored in an array of found rectangles, which is then processed to filter out rectangles of the wrong aspect ratio. Finally, the most prevalent concentric group of rectangles is located by rejecting rectangles whose geometric center is too far from the center of mass of the entire group of located rectangles. This final group of concentric rectangles is then used to automatically target a robotic shooting mechanism or to provide feedback to robot operators for manual targeting.

## Claims

### Claim 1

A method for detecting at least one target rectangle in an image, the method comprising:
receiving, in an image memory, an image from an image source;
transforming the received image into a grayscale edge detection image;
transforming the grayscale edge detection image into a binary image using a threshold value;
determining horizontal and vertical sub-exlons for each picture element in the binary image;
generating horizontal and vertical sub-exlon arrays from the determined horizontal and vertical sub-exlons;
scanning the sub-exlon arrays to identify exlons for each of four corner types; and
identifying the at least one target rectangle using the identified exlons by confirming that one of each of four corner types are properly oriented and connected by correctly oriented horizontal and vertical lines.

### Claim 2 (depends on 1)

The method of claim 1, further comprising
identifying a plurality of target rectangles;
storing a list of the plurality of target rectangles; and
removing all target rectangles from the target rectangle list whose aspect ratios are not substantially consistent with an aspect ratio of a predetermined target rectangle, thus generating a filtered rectangle list.

### Claim 3 (depends on 2)

The method of claim 2, further comprising
determining a set of concentric target rectangles by repeatedly removing from the list of target rectangles those target rectangles whose geometric centers are farthest away from the average of the geometric centers of the concentric target rectangles, this process being repeated until all remaining target rectangles have geometric centers closer than a threshold to average of the geometric centers.

### Claim 4 (depends on 3)

The method of claim 3, further comprising
generating aiming signals in order to align a shooting mechanism with the set of concentric target rectangles remaining in the list; and
transmitting the aiming signals to an external aiming device.

### Claim 5 (depends on 4)

The method of claim 4, wherein the external aiming device is a set of actuators that automatically align the shooting mechanism with the concentric target rectangles.

### Claim 6 (depends on 4)

The method of claim 4, wherein the external aiming device is a set of signal lights that indicate to a human driver which direction a robot must be positioned in order to align with the concentric target rectangles.

### Claim 7 (depends on 1)

The method of claim 1, wherein the image source is a camera.

### Claim 8 (depends on 1)

The method of claim 1, wherein the image source is a computer file.

### Claim 9

A system for aiming a robotic apparatus comprising:
an image source;
a memory that includes instructions for operating the system;
an aiming device; and
control circuitry coupled to the image source, coupled to the memory and coupled to the aiming device, the control circuitry capable of executing the instructions and is operable to at least:
receive an image from an image source and store the image in the memory;
transform the received image into a grayscale edge detection image;
transform the grayscale edge detection image into a binary image using a threshold value;
determine horizontal and vertical sub-exlons for each picture element in the binary image;
generate horizontal and vertical sub-exlon arrays from the determined horizontal and vertical sub-exlons;
scan the sub-exlon arrays to identify exlons for each of four corner types;
identify at least one target rectangle using the identified exlons by confirming that one of each of four corner types are properly oriented and connected by correctly oriented horizontal and vertical lines;
generate aiming signals in order to align a shooting mechanism with the at least one target rectangle; and
transmit the aiming signals to the aiming device.

### Claim 10 (depends on 9)

The system of claim 9, wherein the control circuitry is further operable to execute the instructions to:
identify a plurality of target rectangles;
store a list of the plurality of target rectangles; and
remove all target rectangles from the target rectangle list whose aspect ratios are not substantially consistent with an aspect ratio of a predetermined target rectangle, thus generating a filtered rectangle list.

### Claim 11 (depends on 10)

The system of claim 10, wherein the control circuitry is further operable to execute the instructions to:
determine a set of concentric target rectangles by repeatedly removing from the list of target rectangles those target rectangles whose geometric centers are farthest away from the average of the geometric centers of the concentric target rectangles, this process being repeated until all remaining target rectangles have geometric centers closer than a threshold to average of the geometric centers.

### Claim 12 (depends on 11)

The system of claim 11, wherein the aiming signals are generated in order to align a shooting mechanism with the set of concentric target rectangles remaining in the list.

### Claim 13 (depends on 11)

The system of claim 11, wherein the aiming device is a set of actuators that automatically align the shooting mechanism with the concentric target rectangles.

### Claim 14 (depends on 11)

The system of claim 11, wherein the aiming device is a set of signal lights that indicate to a human driver which direction the robotic apparatus must be positioned in order to align with the concentric target rectangles.

### Claim 15 (depends on 9)

The system of claim 9, wherein the image source is a camera.

### Claim 16 (depends on 9)

The system of claim 9, wherein the image source is a computer file.

### Claim 17

A non-transitory computer-readable medium comprising a plurality of instructions that, when executed by a computer system, at least cause the computer system receive an image from an image source and store the image in the memory;
transform the received image into a grayscale edge detection image;
transform the grayscale edge detection image into a binary image using a threshold value;
determine horizontal and vertical sub-exlons for each picture element in the binary image;
generate horizontal and vertical sub-exlon arrays from the determined horizontal and vertical sub-exlons;
scan the sub-exlon arrays to identify exlons for each of four corner types;
identify at least one target rectangle using the identified exlons by confirming that one of each of four corner types are properly oriented and connected by correctly oriented horizontal and vertical lines;
generate aiming signals in order to align a shooting mechanism with the at least one target rectangle; and
transmit the aiming signals to an external aiming device

### Claim 18 (depends on 17)

The non-transitory computer-readable medium of claim 17, wherein the instructions further cause the computer system to:
identify a plurality of target rectangles;
store a list of the plurality of target rectangles; and
remove all target rectangles from the target rectangle list whose aspect ratios are not substantially consistent with an aspect ratio of a predetermined target rectangle, thus generating a filtered rectangle list.

### Claim 19 (depends on 18)

The non-transitory computer-readable medium of claim 18, wherein the instructions further cause the computer system to:
determine a set of concentric target rectangles by repeatedly removing from the list of target rectangles those target rectangles whose geometric centers are farthest away from the average of the geometric centers of the concentric target rectangles, this process being repeated until all remaining target rectangles have geometric centers closer than a threshold to average of the geometric centers.

### Claim 20 (depends on 19)

The non-transitory computer-readable medium of claim 19, wherein the aiming signals are generated in order to align a shooting mechanism with the set of concentric target rectangles remaining in the list.
