# US 2015/0146918 A1 · Video device for realtime pedaling frequency estimation > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2015/0146918 A1 - Document ID: US20150146918A1 - Status: Published application - Kind: A1 - Application number: 14/255,604 - Filed: 2014-04-17 - Published: 2015-05-28 - Earliest claimed: 2013-11-26 - Assignees: National Chung Cheng University - Applicants: National Chung Cheng University - Inventors: Tay-Jyi LIN; Ching-Wei YEH; Yuan-Hsiang Miao; Shau-Chian TANG - Web page: https://patentgrep.com/patent/US20150146918A1 ## CPC - G06V 20/40: in video content (extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44) - A61B 5/11: Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb {(A61B5/1038 takes precedence; motion detection to correct for motion artifacts in physiological signals A61B5/721)} - A61B 5/1128: {using image analysis (A61B5/1127 takes precedence)} - G06T 7/20: Analysis of motion (motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51) - G06V 20/56: exterior to a vehicle by using sensors mounted on the vehicle - G06V 40/23: {Recognition of whole body movements, e.g. for sport training} - H04N 7/18: Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast - G06T 2207/30221: Sports video; Sports image - G06V 40/25: {Recognition of walking or running movements, e.g. gait recognition} - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract A video device for realtime pedaling frequency estimation is mounted on a bike and comprises an image capture unit capturing continuous dynamic images of an upper body of a biker; an image recognition unit recognizing images of symmetric regions of the biker and images of swings of the symmetric regions from the continuous dynamic images; a microprocessor calculating a frequency of periodical swings of the biker from the images of the symmetric regions and the images of the swings of the symmetric regions, and then obtaining a pedaling frequency; and a display device presenting the pedaling frequency, wherein a widely-used intelligent handheld device replaces the sensors, display device, and complicated circuits of the conventional cyclometer, and wherein a novel image recognition technology is used to measure the pedaling frequency with a considerable accuracy in a lower cost and a convenient way. ## Claims ### Claim 1 A video device for realtime pedaling frequency estimation, mounted on a bike and comprising an image capture unit capturing continuous dynamic images of an upper body of a biker; an image recognition unit electrically connected with said image capture unit and recognizing images of symmetric regions of said biker and images of swings of said symmetric regions from said continuous dynamic images; a microprocessor electrically connected with said image recognition unit, calculating a frequency of periodical swings of said biker from said images of said symmetric regions and said images of said swings of said symmetric regions, and then obtaining a pedaling frequency; and a display device electrically connected with said microprocessor and presenting information of said pedaling frequency. ### Claim 2 (depends on 1) The video device for realtime pedaling frequency estimation according to claim 1, wherein said microprocessor uses said pedaling frequency, a gear ratio of said bike, and a wheel diameter of said bike to work out speed of said bike. ### Claim 3 (depends on 1) The video device for realtime pedaling frequency estimation according to claim 1, wherein said images of said symmetric regions are a left shoulder image and a right shoulder image of said biker, and wherein a first reference point and a second reference point are respectively established at said left shoulder image and said right shoulder image of said biker, and wherein images of continuous variation of a relative tilt angle of said first reference point and said second reference point are used as said images of said swings of said symmetric regions. ### Claim 4 (depends on 3) The video device for realtime pedaling frequency estimation according to claim 3, wherein said image recognition unit uses an edge detection method to detect edge pixels having dramatic gradient variation in said continuous dynamic images and uses a Hough Transform to connect said edge pixels to define a left shoulder line of said left shoulder image and a right shoulder line of said right shoulder image. ### Claim 5 (depends on 3) The video device for realtime pedaling frequency estimation according to claim 3, wherein said microprocessor uses a period of swings of said left shoulder line and said right shoulder line and variation of a relative tilt angle of said left shoulder line and said right shoulder line to calculate said frequency of said periodical swings of said hiker, and then obtains said pedaling frequency. ### Claim 6 (depends on 3) The video device for realtime pedaling frequency estimation according to claim 3, wherein an intersection of said left shoulder line and a neck of said biker is used as a start point of said left shoulder line, and said first reference point is defined via extending said start point of said left shoulder line outward by 50 pixels, and wherein an intersection of said right shoulder line and said neck of said biker and is used as a start point of said right shoulder line, and said second reference point is defined via extending said start point of said right shoulder line outward by 50 pixels. ### Claim 7 (depends on 1) The video device for realtime pedaling frequency estimation according to claim 1, wherein said images of said symmetric regions are a left ear image and a right ear image of said biker, and wherein a first reference point and a second reference point are respectively established at said left ear image and said right ear image of said biker, and wherein images of continuous variation of a relative tilt angle of said first reference point and said second reference point are used as said images of said swings of said symmetric regions. ### Claim 8 (depends on 7) The video device for realtime pedaling frequency estimation according to claim 7, wherein said image recognition unit uses an edge detection method to detect edge pixels having dramatic gradient variation in said continuous dynamic images and uses a Hough Transform to connect said edge pixels to define a left ear contour line of said left ear image and a right ear contour line of said right ear image. ### Claim 9 (depends on 7) The video device for realtime pedaling frequency estimation according to claim 7, wherein said microprocessor uses a period of swings of said left ear contour line and said right ear contour line and variation of a relative tilt angle of said left ear contour line and said right ear contour line to calculate said frequency of said periodical swings of said biker, and then obtains said pedaling frequency. ### Claim 10 (depends on 1) The video device for realtime pedaling frequency estimation according to claim 1, wherein said image capture unit is CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor) sensor. ### Claim 11 (depends on 1) The video device for realtime pedaling frequency estimation according to claim 1, wherein said image recognition unit is a human image recognition program built in said microprocessor.