KR-20260063285-A - Method and system for surveillance camera calibration based on wireless GPS transceiver for estimating position of detected object

KR20260063285AKR 20260063285 AKR20260063285 AKR 20260063285AKR-20260063285-A

Abstract

A method and apparatus are provided for calibrating a surveillance camera device based on a wireless GPS transceiver and estimating the location of a detected object using the calibrated camera attitude information. A method according to one embodiment of the present invention comprises: an information collection step of acquiring actual location information of at least four structures using a wireless GPS transceiver from an image captured by a surveillance camera device, storing pixel location information, actual location information, and partial images within the image of each structure, and calculating camera attitude information; and a camera correction step of correcting the camera attitude information using a newly captured image and the pixel location information, actual location information, and partial images within the image of each structure stored in the information collection step.

Inventors

김휘

Assignees

한국전자통신연구원

Dates

Publication Date: 20260507
Application Date: 20241030

Claims (20)

Information collection step of acquiring actual location information of at least four structures using a wireless GPS transceiver from an image captured by a surveillance camera device, storing pixel location information, actual location information, and partial images within the image of each structure, and calculating camera attitude information; and A camera calibration step that corrects camera attitude information using a newly captured image and pixel position information, actual position information, and partial images within the image of each structure stored in the information collection step. A surveillance camera calibration method comprising
In paragraph 1, the information collection step is, A first step of capturing an image from a surveillance camera device and transmitting the captured image to a wireless GPS transceiver; A second step in which, for each of the at least four structures, when a bottom center point is selected within the image, a wireless GPS transceiver measures the actual location and transmits pixel location information within the image and actual location information of the bottom center point of each structure to a surveillance camera device; A third step in which a surveillance camera device stores pixel location information within the image of the bottom center point of each structure, actual location information, and partial images; A fourth step of calculating camera attitude information using pixel position information within the image and actual position information of the bottom center point of each of the above structures. A surveillance camera calibration method including
In paragraph 2, the second stage is, For each of the above at least four structures The above wireless GPS transceiver receives a selection of the bottom center point of the structure in the above image, and A step in which, when the transmit button is pressed, the wireless GPS transceiver measures the current location as the actual location of the structure, and A wireless GPS transceiver transmitting pixel location information within an image of the bottom center point of the structure and the measured actual location information to a surveillance camera device. A surveillance camera calibration method that performs...
In paragraph 3, the above second step is, The step of the wireless GPS transceiver receiving the image from the surveillance camera device displaying the image, and The step of the wireless GPS transceiver selecting at least four structures in the indicated image A surveillance camera calibration method including further
In paragraph 4, When a wireless GPS transceiver transmits pixel location information within an image of the bottom center point of the structure and the measured actual location information to a surveillance camera device, it also transmits information regarding a partial image of the structure. Surveillance camera calibration method.
In paragraph 3, the above first step is, A surveillance camera device detects and selects at least four structural objects in the captured image, and transmits an image showing the selected at least four structures to a wireless GPS transceiver. Surveillance camera calibration method.
In paragraph 2, in the above fourth step, Let the pixel location information of the bottom center point of each structure be (x p , y p ), and the actual location information be (X W , Y W , Z W ), where (f x , f y ) is the camera focal length and (c x , c y ) is the principal point, which is a pre-given camera feature, The camera's attitude information ([R|t]) is A surveillance camera calibration method obtained by
In paragraph 2, the camera correction step is, Step 5, in which the surveillance camera device captures an image; Step 6: finding a part that matches the partial image of the at least four structures stored in the information collection step in the image taken in Step 5; Step 7: acquiring pixel location information within the image taken in Step 5 of the lower center point of the matched part of the at least four structures; Step 8, recalculating camera attitude information using pixel location information acquired in Step 7 of the above at least 4 structures and actual location information stored in the information collection step. A surveillance camera calibration method including
In any one of paragraphs 1 through 8, An object position estimation step that detects an object and estimates the actual position of the object based on the bottom center point of the object using camera pose information regenerated in the camera calibration step. A surveillance camera calibration method further equipped with
In Clause 9, in the object location estimation step, Let the coordinates of the bottom center point of the object be (x p , y p ), the camera pose information be [R|t], and the actual position of the object be (X W , Y W , Z W ). Then the following equation A surveillance camera calibration method that estimates the actual location of an object.
A surveillance camera device comprising a camera for capturing images and a first wireless communication unit for wirelessly communicating with an external device; and A wireless GPS transceiver comprising a second wireless communication unit for wirelessly communicating with the surveillance camera device, a GPS sensor for acquiring current location information, and a user interface unit, receiving an image captured by the surveillance camera device, acquiring actual location information of at least four structures included in the image, and transmitting pixel location information within the image and actual location information of each structure to the surveillance camera device. Equipped with, The above surveillance camera device stores pixel location information and actual location information within an image of each structure, and calculates camera attitude information using the above. Surveillance camera calibration system.
In Paragraph 11, The above surveillance camera device further comprises a camera correction unit, and The above surveillance camera device stores partial images of the at least four structures, and The above surveillance camera device periodically captures images for correction, and The above camera correction unit corrects camera attitude information using the image for correction, pixel position information within the image of each structure stored above, actual position information, and partial images. Surveillance camera calibration system.
In Paragraph 12, In information collection mode, the surveillance camera device captures an image and transmits the captured image to a wireless GPS transceiver, and When a bottom center point is selected for each of the at least four structures within the captured image, the wireless GPS transceiver measures the actual location of the structure and transmits the pixel location information within the image and the actual location information of the bottom center point of each structure to the surveillance camera device. A surveillance camera device stores pixel position information within an image of the bottom center point of each structure, actual position information, and a partial image, and calculates camera attitude information using the pixel position information within the image of the bottom center point of each structure and the actual position information. Surveillance camera calibration system.
In Clause 13, the selection of the above-mentioned at least four structures is, A wireless GPS transceiver that receives an image from a surveillance camera device displays the received image on the user interface unit, and receives the selection of at least four structures from the user in the displayed image, Surveillance camera calibration system.
In Paragraph 14, When a wireless GPS transceiver transmits pixel location information within an image of the bottom center point of the structure and the measured actual location information to a surveillance camera device, it also transmits information regarding a partial image of the structure. Surveillance camera calibration system.
In Clause 13, the selection of the above-mentioned at least four structures is, The surveillance camera device detects and selects at least four structural objects from the captured image, and The surveillance camera device transmits an image showing at least four selected structures to a wireless GPS transceiver, Surveillance camera calibration system.
In Paragraph 13, Let the pixel location information of the bottom center point of each structure be (x p , y p ), and the actual location information be (X W , Y W , Z W ), where (f x , f y ) is the camera focal length and (c x , c y ) is the principal point, which is a pre-given camera feature, The attitude information ([R|t]) of the above camera is A surveillance camera calibration system obtained by
In Clause 12, the camera correction unit is, Finding a part that matches the partial image of the at least four structures in the image for correction, obtaining pixel location information within the image for correction of the bottom center point of the matched part of the at least four structures, and recalculating camera pose information using the obtained pixel location information of the at least four structures and the stored actual location information. Surveillance camera calibration system.
In any one of paragraphs 11 to 18, the surveillance camera device is, An object detection unit for detecting objects in a captured image, and A position estimation unit that estimates the actual position of the object based on the bottom center point of the object using the above camera pose information. A surveillance camera calibration system that further includes
In Clause 19, the above-mentioned position estimation unit, Let the coordinates of the bottom center point of the object be (x p , y p ), let the camera pose information be [R|t], and let the actual position of the object be (X W , Y W , Z W ), A surveillance camera calibration system that estimates the actual location of an object.

Description

Method and system for surveillance camera calibration based on wireless GPS transceiver for estimating the position of a detected object The present invention relates to a method and apparatus for calibrating a surveillance camera device based on a wireless GPS transceiver and estimating the position of a detected object using calibrated camera attitude information. Surveillance cameras are widely installed for purposes such as crime prevention, traffic enforcement, and intelligence gathering. Traditionally, estimating the actual location (latitude and longitude) of objects captured by surveillance cameras required a significant amount of prior information, including the camera's installation position (latitude, longitude, and altitude) and camera attitude data (pan-tilt, rotation angle). Furthermore, precise location estimation necessitates accurate installation and configuration of the surveillance cameras. Consequently, substantial costs and time are invested in training field personnel, or system developers are often deployed. Furthermore, after the installation of surveillance cameras, information such as camera settings (pan-tilt, rotation angle) may change due to surrounding environmental factors (road vibrations, wind, etc.). Depending on camera characteristics such as the field of view, even slight changes in rotation can significantly reduce the accuracy of object location estimation within the camera image. FIG. 1 is a configuration diagram of a surveillance camera calibration system according to one embodiment of the present invention. Figure 2 is a flowchart showing the information collection procedure for camera calibration. Figure 3 is an explanatory diagram for describing the information collection procedure. Figure 4 is an explanatory diagram for describing a method to estimate the actual location of an object. Figure 5 shows the difference in images taken when the camera position is changed by the surrounding environment. Figure 6 is a flowchart showing the procedure for correcting camera attitude information. Figure 7 is an explanatory diagram for explaining the principle of correcting camera attitude information through partial image matching when the camera attitude is changed by the surrounding environment. The aforementioned objectives of the present invention, as well as other objectives, advantages, and features, and the methods for achieving them, will become clear from the embodiments described in detail below together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but can be implemented in various different forms, and the following embodiments are provided merely to easily inform those skilled in the art of the purpose, structure, and effects of the invention, and the scope of the rights of the present invention is defined by the description in the claims. Meanwhile, the terms used in this specification are for describing the embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used in this specification, "comprises" and/or "comprising" do not exclude the presence or addition of one or more other components, steps, actions, and/or elements to the mentioned components, steps, actions, and/or elements. FIG. 1 is a configuration diagram of a surveillance camera calibration system according to one embodiment of the present invention. A camera calibration system based on a wireless GPS transceiver for estimating the location of a detected object according to one embodiment of the present invention includes a surveillance camera device (100) and a wireless GPS transceiver (200). A surveillance camera device (100) includes a camera (110), an object detection unit (120) for detecting an object in an image captured by the camera, a position estimation unit (130) for estimating the position of the detected object, a camera correction unit (140) for correcting the pose information of the camera, and a wireless communication unit (150) for wireless communication with a wireless GPS transceiver (200). A wireless GPS transceiver (200) may include a wireless communication unit (210) for wireless communication with a surveillance camera device (100), a GPS sensor (220) for acquiring current location information, and a user interface unit (230) including means for instructing the transmission of location information. The user interface unit (230) may include an input means for inputting operation commands and a display means for displaying images. In one embodiment, the wireless GPS transceiver (200) may be implemented in the form of an app installed on a smartphone or tablet device. The surveillance camera device (100) transmits the captured image to the wireless GPS transceiver (200), and the wireless GPS transceiver (200) obtains location information of road structures or accessories (hereinafter referred to