IP Library › Granted Patent US 11,842,634
Granted Patent B2
US 11,842,634 · App. 17/422,980 · Granted Dec 12, 2023

Image detection device, signal control system compromising same and signal control method

Inventors: Tae Wook Choi (Seongnam-si, KR); Jung Woo Hwang (Seoul, KR); Kyungjean Bae (Seoul, KR); Hyun Woo Cho (Seongnam-si, KR); Gun Woo Jang (Seongnam-si, KR); Hee Bin Lee (Incheon, KR)
Assignee: LAON ROAD INC.
G08G1/07G06N3/02G06T7/70G06V20/54G06T2207/20076G06T2207/20084G06T2207/30236G06V2201/08
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,842,634
App. No.
17/422,980
Granted
Dec 12, 2023
Kind
B2
Abstract

Proposed herein are an image detection device, a signal control system including the same, and a signal control method. The image detection device includes: a photographing unit configured to acquire an intersection image by capturing an image of an intersection; a control unit configured to generate detection information, including information about a vehicle-occupied or -unoccupied state for each of one or more detection channels set on the plane of the intersection by analyzing the intersection image acquired by the photographing unit; and a communication unit configured to transmit the detection information, generated by the control unit, to the traffic signal controller.

Claims (45)

1. A signal control system comprising:

an image detection device for providing detection information to a traffic signal controller; and

the traffic signal controller configured to receive the detection information from the image detection device, and to generate control information about a signal to be flickered in traffic lights based on the received detection information,

wherein the image detection device comprises:

a photographing unit configured to acquire an intersection image by capturing an image of an intersection;

a control unit configured to generate the detection information, including information about a vehicle-occupied or -unoccupied state for each of one or more detection channels set on a plane of the intersection by analyzing the intersection image acquired by the photographing unit; and

a communication unit configured to transmit the detection information, generated by the control unit, to the traffic signal controller,

wherein the traffic signal controller comprises:

at least one option board configured to receive the detection information from the image detection device, and to generate detection data by converting the detection information into the detection data including 4-bit channel information for each detection channel;

a main board configured to receive the detection data from the option board, and to generate a control signal for actuated control or spillback control of the intersection; and

a signal drive unit configured to receive the control signal from the main board, to generate a drive signal for the traffic lights according to the control signal, and to transfer the generated drive signal to the traffic lights,

wherein when generating the detection data, the option board writes information, indicating a vehicle-occupied or -unoccupied state for the detection channel included in the detection information, in one bit included in the channel information and writes information indicating a state of the image detection device in remaining three bits, and

wherein channel information for four or eight detection channels is included in one piece of the detection data.

2. The signal control system of claim 1 , wherein the control unit acquires location coordinates of each object within the intersection image by using an artificial neural network that receives an intersection image as an input, recognizes an object included in the intersection image and estimated to be a vehicle, and outputs information on a location of the recognized object, converts the acquired location coordinates of each object within the intersection image into location coordinates on the plane of the intersection, and then determines the detection channel to be in an occupied state when the resulting location coordinates are included in the detection channel.

3. The signal control system of claim 1 , wherein the control unit extracts pixels corresponding to a vehicle by using an artificial neural network that receives an intersection image as an input and outputs a probability map indicating a probability that each pixel included in the intersection image corresponds to a vehicle, converts each of the extracted pixels into a pixel on the plane of the intersection, and determines a vehicle-occupied or -unoccupied state for the detection channel according to a number of resulting pixels included in the detection channel.

4. The signal control system of claim 1 , wherein the detection channel is set for each lane of a stop line of an entry to the intersection, and is set for each lane or each road of a standby line spaced apart backward from the stop line by a preset distance.

5. The signal control system of claim 1 , wherein the detection channel is set for an exit from the intersection or for each lane of the exit.

6. The signal control system of claim 1 , further comprising:

the traffic lights configured to be driven under a control of the traffic signal controller.

7. The signal control system of claim 1 , wherein:

the image detection device generates the detection information at a period corresponding to an integer multiple of a period at which the option board transfers the detection data to the main board, and transfers the detection information to the option board; and

the option board repeatedly generates detection data, generated using the detection information received from the image detection device, a number of times equal to the integer multiple.

8. The signal control system of claim 1 , wherein:

the image detection device differentiates a total of 64 detection channels, and generates the detection information for each of the detection channels; and

the option board is matched to every eight detection channels.

9. A signal control method performed by a signal control system including an image detection device and a traffic signal controller, the signal control method comprising:

acquiring an intersection image by capturing an image of an intersection;

generating detection information, including information about a vehicle-occupied or -unoccupied state for each of one or more detection channels set on a plane of the intersection by analyzing the intersection image;

generating detection data by converting the detection information into 4-bit channel information for each of the detection channels;

generating a control signal for actuated control or spillback control of the intersection based on the detection data; and

transferring a drive signal to traffic lights based on the control signal,

wherein generating the detection data comprises writing information, indicating a vehicle-occupied or -unoccupied state for the detection channel included in the detection information, in one bit included in the channel information and writing information indicating a state of the image detection device in remaining three bits, and

wherein channel information for four or eight detection channels is included in one piece of the detection data.

10. The signal control method of claim 9 , wherein generating the detection information comprises:

acquiring location coordinates of each object within the intersection image by using an artificial neural network that receives an intersection image as an input, recognizes an object included in the intersection image and estimated to be a vehicle, and outputs information on a location of the recognized object;

converting the acquired location coordinates of each object within the intersection image into location coordinates on the plane of the intersection;

determining the detection channel to be in an occupied state when the resulting location coordinates are included in the detection channel, and determining the detection channel to be in an unoccupied state when the resulting location coordinates are not included in the detection channel; and

generating the detection information according to the determined state.

11. The signal control method of claim 9 , wherein generating the detection information comprises:

extracting pixels corresponding to a vehicle by using an artificial neural network that receives an intersection image as an input and outputs a probability map indicating a probability that each pixel included in the intersection image corresponds to a vehicle;

converting each of the extracted pixels into a pixel on the plane of the intersection;

determining the detection channel to be in an occupied state when a number of resulting pixels included in the detection channel is equal to or larger than a preset number, and determining the detection channel to be in an unoccupied state when the number of resulting pixels included in the detection channel is smaller than the preset number; and

generating the detection information according to the determined state.

12. The signal control method of claim 9 , wherein generating the detection information comprises repeatedly generating the detection data based on a period at which the option board receives the detection information from the image detection device and a period at which the option board transfers the detection data to the main board.

13. A computer-readable non-transitory storage medium having stored thereon a program for performing the signal control method of claim 9 .

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 5, 2022
From: LAON PEOPLE INC.
To: LAON ROAD INC.
Reel/Frame 059830/0123 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 16, 2021
From: CHOI, TAE WOOK; HWANG, JUNG WOO; BAE, KYUNGJEAN; CHO, HYUN WOO; JANG, GUN WOO; LEE, HEE BIN
To: LAON PEOPLE INC.
Reel/Frame 056898/0292 →
Priority Claims (1)
KR 10-2019-0134923 · Oct 28, 2019 · national
Continuity (1)
Related Publication 20220076569A1 · Mar 10, 2022