CN-121977456-A - Thickness measuring method, device, equipment, medium and product of tread cleaning device
Abstract
The embodiment of the invention provides a thickness measuring method, a device, equipment, a medium and a product of a tread cleaning device, wherein the method comprises the steps of obtaining an image of the tread cleaning device; the method comprises the steps of carrying out semantic segmentation on an image of a tread cleaning device, determining a mask segmentation area of the tread cleaning device in the image of the tread cleaning device, determining a minimum circumscribed rectangle of the tread cleaning device according to the mask segmentation area of the tread cleaning device in the image of the tread cleaning device, and determining the thickness of the tread cleaning device according to the minimum circumscribed rectangle of the tread cleaning device and the length of the tread cleaning device. By utilizing the method, the thickness measurement of the tread cleaning device is realized based on the image segmentation, the mask circumscribed rectangle and the actual distance mapping relation by utilizing the tread cleaning device image, the difficult problem of thickness measurement under the condition of lacking reliable point cloud is solved, the accuracy of the thickness measurement of the tread cleaning device is improved, and the method is stable, reliable and high in feasibility.
Inventors
- LI SHA
- LIU QINJIE
- ZHAO HONGCHANG
Assignees
- 北京铁道工程机电技术研究所股份有限公司
Dates
- Publication Date
- 20260505
- Application Date
- 20260116
Claims (10)
- 1. A method for measuring thickness of a tread cleaning device, comprising: acquiring an image of the tread cleaning device; Performing semantic segmentation on the tread cleaning device image to determine a mask segmentation area of the tread cleaning device in the tread cleaning device image; Determining the minimum circumscribed rectangle of the tread cleaning device according to the mask dividing region of the tread cleaning device in the tread cleaning device image; and determining the thickness of the tread cleaning device according to the minimum circumscribed rectangle of the tread cleaning device and the length of the tread cleaning device.
- 2. The method according to claim 1, wherein said semantically segmenting the tread cleaning device image to determine a masked segmented region of the tread cleaning device in the tread cleaning device image comprises: and performing semantic segmentation on the tread cleaning device image by utilizing a pre-trained image segmentation network, and determining a mask segmentation area of the tread cleaning device in the tread cleaning device image.
- 3. The method according to claim 1, wherein said determining a minimum bounding rectangle of the tread cleaning device from the masked segmented regions of tread cleaning devices in the tread cleaning device image comprises: Determining the inclination angle of an initial circumscribed rectangle of the tread cleaning device in the tread cleaning device image according to a mask dividing region of the tread cleaning device in the tread cleaning device image; According to the inclination angle, performing correction processing on the mask dividing region to obtain a corrected mask dividing region; And determining the minimum circumscribed rectangle of the tread cleaning device in the tread cleaning device image according to the mask dividing region after the alignment.
- 4. The method according to claim 3, further comprising, before said determining an inclination angle of an initial circumscribed rectangle of the tread cleaning device in the tread cleaning device image from the masked segmented region of the tread cleaning device in the tread cleaning device image: and denoising and downsampling the mask segmentation area of the tread cleaning device in the tread cleaning device image to obtain the processed mask segmentation area.
- 5. The method of claim 1, wherein the determining the thickness of the tread cleaning device based on the minimum bounding rectangle of the tread cleaning device and the length of the tread cleaning device comprises: Determining the length-width ratio of the tread cleaning device according to four corner points of the minimum circumscribed rectangle of the tread cleaning device; Acquiring the length of the tread cleaning device; And determining the thickness of the tread cleaning device according to the length-width ratio of the tread cleaning device and the length.
- 6. The method according to claim 1, wherein said acquiring tread cleaning device images comprises: acquiring a captured image related to a train by an image acquisition device; screening candidate images containing the tread cleaning device from the captured images according to the image identifications of the tread cleaning device images; Processing the candidate images by adopting a template matching technology, and determining the position of the tread cleaning device in the candidate images; and expanding a set number of pixels to the periphery according to the position of the tread cleaning device in the candidate image to cut the candidate image, so as to obtain the tread cleaning device image.
- 7. A thickness measuring device for a tread cleaning device, comprising: The image acquisition module is used for acquiring images of the tread cleaning device; the semantic segmentation module is used for carrying out semantic segmentation on the tread cleaning device image and determining a mask segmentation area of the tread cleaning device in the tread cleaning device image; The rectangle determining module is used for determining the minimum circumscribed rectangle of the tread cleaning device according to the mask dividing region of the tread cleaning device in the tread cleaning device image; And the thickness determining module is used for determining the thickness of the tread cleaning device according to the minimum circumscribed rectangle of the tread cleaning device and the length of the tread cleaning device.
- 8. An electronic device, comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of thickness measurement of the tread cleaning device of any one of claims 1-6.
- 9. A computer readable storage medium storing computer instructions for causing a processor to perform the method of thickness measurement of the tread cleaning device according to any one of claims 1-6.
- 10. A computer program product, characterized in that it comprises a computer program which, when executed by a processor, implements the method for thickness measurement of a tread cleaning device according to any one of claims 1 to 6.
Description
Thickness measuring method, device, equipment, medium and product of tread cleaning device Technical Field The invention relates to the technical field of railway vehicles, in particular to a thickness measuring method, a thickness measuring device, thickness measuring equipment, thickness measuring media and thickness measuring products of a tread cleaning device. Background The tread cleaning device can be contacted with the tread of the wheel when being opened, and impurities on the tread of the wheel are removed through friction, so that the tread of the wheel is cleaned, and the normal work of a train is ensured. The tread cleaning device is worn and thinned along with friction, and the thickness state is a key index for judging the wear degree and the residual life. In the thickness detection of a high-speed railway tread sweeping device, the traditional method relying on 3D point cloud has the limitation that the three-dimensional camera of the inspection robot cannot capture complete and clear point cloud data because the device is positioned on the upper layer of a wheel set, so that thickness measurement cannot be realized based on the point cloud data. Disclosure of Invention The embodiment of the invention provides a thickness measurement method, a device, equipment, a medium and a product of a tread cleaning device, which realize the thickness measurement of the tread cleaning device, solve the difficult problem of thickness measurement under the condition of lacking a reliable point cloud and improve the accuracy of the thickness measurement of the tread cleaning device. In a first aspect, the present embodiments provide a method for measuring thickness of a tread cleaning device, the method comprising: acquiring an image of the tread cleaning device; Performing semantic segmentation on the tread cleaning device image to determine a mask segmentation area of the tread cleaning device in the tread cleaning device image; Determining the minimum circumscribed rectangle of the tread cleaning device according to the mask dividing region of the tread cleaning device in the tread cleaning device image; and determining the thickness of the tread cleaning device according to the minimum circumscribed rectangle of the tread cleaning device and the length of the tread cleaning device. In a second aspect, the present embodiments provide a thickness measuring device for a tread cleaning device, the device comprising: The image acquisition module is used for acquiring images of the tread cleaning device; the semantic segmentation module is used for carrying out semantic segmentation on the tread cleaning device image and determining a mask segmentation area of the tread cleaning device in the tread cleaning device image; The rectangle determining module is used for determining the minimum circumscribed rectangle of the tread cleaning device according to the mask dividing region of the tread cleaning device in the tread cleaning device image; And the thickness determining module is used for determining the thickness of the tread cleaning device according to the minimum circumscribed rectangle of the tread cleaning device and the length of the tread cleaning device. In a third aspect, the present embodiment provides an electronic device, including: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of thickness measurement of a tread cleaning device according to any of the embodiments of the present invention. In a fourth aspect, the present embodiment provides a computer-readable storage medium storing computer instructions for causing a processor to perform a method for thickness measurement of a tread cleaning device according to any embodiment of the present invention. In a fifth aspect, embodiments of the present invention also provide a computer program product comprising a computer program which, when executed by a processor, implements a method of thickness measurement for a tread cleaning device according to any embodiment of the present invention. The embodiment of the invention provides a thickness measuring method, a device, equipment, a medium and a product of a tread cleaning device, wherein the method comprises the steps of firstly acquiring an image of the tread cleaning device; the method comprises the steps of carrying out semantic segmentation on an image of a tread cleaning device, determining a mask segmentation area of the tread cleaning device in the image of the tread cleaning device, determining a minimum circumscribed rectangle of the tread cleaning device according to the mask segmentation area of the tread cleaning device in the image of the tread cleaning device, and finally determining the thickness of the tread cleaning device according to the minimum circumscribed rectangle of the tread cleanin