CN112818842A - Intelligent image recognition swimming timing system and timing method based on machine learning - Google Patents

Intelligent image recognition swimming timing system and timing method based on machine learning Download PDF

Info

Publication number
CN112818842A
CN112818842A CN202110129228.0A CN202110129228A CN112818842A CN 112818842 A CN112818842 A CN 112818842A CN 202110129228 A CN202110129228 A CN 202110129228A CN 112818842 A CN112818842 A CN 112818842A
Authority
CN
China
Prior art keywords
image
power supply
lane
terminal
swimming
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110129228.0A
Other languages
Chinese (zh)
Other versions
CN112818842B (en
Inventor
徐文海
徐汉隆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN202110129228.0A priority Critical patent/CN112818842B/en
Publication of CN112818842A publication Critical patent/CN112818842A/en
Application granted granted Critical
Publication of CN112818842B publication Critical patent/CN112818842B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • A—HUMAN NECESSITIES
    • A63—SPORTS; GAMES; AMUSEMENTS
    • A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
    • A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
    • A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
    • A63B71/0686—Timers, rhythm indicators or pacing apparatus using electric or electronic means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/254—Analysis of motion involving subtraction of images
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
    • G07C1/22—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people in connection with sports or games
    • G07C1/24—Race time-recorders
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20224—Image subtraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Psychiatry (AREA)
  • Social Psychology (AREA)
  • Human Computer Interaction (AREA)
  • Physical Education & Sports Medicine (AREA)
  • Image Processing (AREA)

Abstract

The invention provides an intelligent image recognition swimming timing system and a timing method based on machine learning, the system comprises a swimming pool, wherein K lanes are arranged in the swimming pool, the K is a positive integer larger than or equal to 1 and is respectively a 1 st lane, a 2 nd lane, a 3 rd lane, … … and a K th lane, and the intelligent image recognition swimming timing system is characterized in that an image acquisition display and a timer connected with the image acquisition display are arranged at the starting end or/and the opposite bank end of each lane; the system also comprises a timing host, a score processing host and a large screen display, wherein the timing host is respectively connected with the score processing host and each timer, and the score processing host is connected with the large screen; and displaying the personal information and the corresponding achievement of the swimmer in each lane on a large screen. The invention can time swimmers in the swimming lane according to the camera.

Description

Intelligent image recognition swimming timing system and timing method based on machine learning
Technical Field
The invention relates to the technical field of image processing, in particular to an intelligent image recognition swimming timing system and method based on machine learning.
Background
Swimming sports are widely favored by people, and swimming and water playing in a swimming pool are good methods for people to relieve summer heat and relax in summer. In swimming projects such as free swimming, breaststroke, backstroke and butterfly swimming, a swimmer is generally prescribed to prepare for launching on a swimming pool starting platform, and in a preparation stage, the swimmer gets on the swimming pool starting platform, steps on the swimming pool starting platform with both feet, jumps into water from the swimming pool starting platform after a command signal is sent, finishes entering water, and generates a final score by flapping a touch panel when reaching a terminal point. Patent application No. 2019110762818 entitled "a swimming pool starting platform, swimming training system and training method", includes: a starting platform body arranged at one side of the swimming pool; the human body induction sensor and the display screen are arranged on the starting table body; a controller; and a timer; when the swimmer contacts with the table board of the starting table body, the human body induction sensor induces the contact of a human body to generate human body identification information, when the swimmer is separated from the table board of the starting table body, the controller responds to the fact that the human body induction sensor does not generate the human body induction information to control the timer to start timing, and the controller is further used for displaying the human body identification information on the display screen. The invention realizes the purpose of automatically timing the swimming time of the swimmer.
Disclosure of Invention
The invention aims to at least solve the technical problems in the prior art, and particularly provides an intelligent image recognition swimming timing system and a timing method based on machine learning.
In order to achieve the above purpose, the invention provides an intelligent image recognition swimming timing system based on machine learning, which comprises a swimming pool, wherein K lanes are arranged in the swimming pool, the K is a positive integer greater than or equal to 1 and is respectively a 1 st lane, a 2 nd lane, a 3 rd lane, … … and a K th lane, and an image acquisition display and a timer connected with the image acquisition display are arranged at the starting end or/and the opposite end of each lane;
the system comprises a score processing host, a large screen, a plurality of timers, a timing host, a score processing host and a display, wherein the score processing host is connected with the large screen; and displaying the personal information and the corresponding achievement of the swimmer in each lane on a large screen.
In a preferred embodiment of the invention, the system further comprises a printer, and the achievement processing host sends the personal information of the swimmers in each lane and the corresponding achievement to the printer for printing;
or/and the swimmer further comprises a mobile terminal, and the mobile terminal is used for checking the personal information and the corresponding achievement of each swimmer in each lane.
In a preferred embodiment of the invention, the device further comprises a starter and a sound box, and the sound box is used for emitting the starting sound of the starter.
In a preferred embodiment of the present invention, K is 10, that is, 10 lanes are provided in the swimming pool, which are lane 1, lane 2, lane 3, lane … … and lane 10, and an image capture display and a timer connected to the image capture display are provided at the beginning and the opposite end of each lane.
In a preferred embodiment of the present invention, the image capturing display includes a column disposed on the ground on the bank side of the lane for supporting a panel, the top end of the column is provided with the panel, the front of the panel is provided with a display screen mounting region for fixedly mounting the display screen, a speaker mounting region for fixedly mounting a speaker, a camera mounting region for mounting a camera, a button mounting region for mounting a button, and a numeric keypad mounting region for mounting a numeric keypad; the display screen is fixedly installed in the display screen installation area, the sound box is fixedly installed in the sound box installation area, the camera is fixedly installed in the camera installation area, the button is fixedly installed in the button installation area, and the numeric keyboard is installed in the numeric keyboard installation area;
a mainboard mounting seat for fixedly mounting a mainboard is arranged in the panel, the mainboard is fixedly mounted on the mainboard mounting seat, and a controller, a timer and a power supply module are arranged on the mainboard; the timing end of the controller is connected with the timing end of the timer, the display data end of the controller is connected with the display data end of the display screen, the sound box data end of the controller is connected with the data sound box end of the sound box, the image data input end of the controller is connected with the image data output end of the camera, the button end of the controller is connected with the data trigger end of the button, and the digital keyboard data end of the controller is connected with the data trigger end of the digital keyboard;
the underwater vehicle driving system also comprises a digital display screen, a water surface camera and an underwater camera which are arranged on the side walls of the lanes on the water surface and under the water, wherein the display data end of the digital display screen is connected with the digital display data end of the controller, and the image data output end of the underwater camera is connected with the underwater image data input end of the controller; the power supply module is respectively connected with a power supply end of the display screen, a power supply end of the sound box, a power supply end of the camera, a power supply end of the button, a power supply end of the numeric keyboard, a power supply end of the digital display screen and a power supply end of the underwater camera, and respectively supplies power to the display screen, the sound box, the camera, the button, the numeric keyboard, the digital display screen and the underwater camera;
and timing swimmers in the swim lane according to the camera.
In a preferred embodiment of the present invention, a power supply module includes: a power supply terminal VCC of the USB interface U1 is respectively connected with a first terminal of a resistor R1, a first terminal of a resistor R2, a first terminal of a resistor R4, an anode of a diode D1, a gate of a field effect transistor Q1 and a power supply terminal VCC of a power management unit U2, the power supply terminal of the USB interface U1 outputs a power supply V _ USB, a ground terminal GND of the USB interface U1 is connected with a power ground, a data positive terminal D +/DP of the USB interface U1 is connected with a USB data positive terminal of the base controller, and a data negative terminal D-/DM of the USB interface U1 is connected with a USB data negative terminal of the base controller; a second end of the resistor R4 is connected with the power ground;
a charging terminal CHRG of the power management unit U2 is respectively connected with a second terminal of the resistor R1 and a negative terminal of the charging indicator light LED1, a positive terminal of the charging indicator light LED1 is connected with a second terminal of the resistor R2, a battery terminal BAT of the power management unit U2 is respectively connected with a positive terminal of the battery holder U3 and a source of the field-effect transistor Q1, a negative terminal of the battery holder U3 is connected with a power ground, a current terminal PROG of the power management unit U2 is connected with a first terminal of the resistor R3, and a second terminal of the resistor R3 is respectively connected with the power ground and a ground terminal GND of the power management unit U2;
a drain electrode of the field effect transistor Q1 and a cathode electrode of the diode D1 are respectively connected with a first end of a capacitor C1, a first end of a capacitor C2, a first end of a resistor R5 and a power supply voltage input end Vin of the voltage reduction unit U4, a drain electrode of the field effect transistor Q1 and a cathode electrode of the diode D1 output a power supply VCC, a second end of the capacitor C1 and a second end of a capacitor C2 are respectively connected with a power supply ground, a second end of the resistor R5 is respectively connected with a first end of a resistor R6 and a first end of the capacitor C3, and a second end of the resistor R6 and a second end of the capacitor C3 are respectively connected with the power supply ground;
a power supply voltage output end Vout of the voltage reduction unit U4 is respectively connected with a first end of a capacitor C4, a first end of a capacitor C5, a first end of a capacitor C6 and a power supply voltage input end Vin of a voltage reduction unit U5, the power supply voltage output end Vout of the voltage reduction unit U4 outputs a power supply 3V3, a ground end GND of the voltage reduction unit U5 is connected with a power ground, and a second end of the capacitor C4, a second end of a capacitor C5 and a second end of the capacitor C6 are respectively connected with the power ground;
a power supply voltage output end Vout of the voltage reduction unit U5 is connected to a first end of the capacitor C7 and a first end of the capacitor C8, respectively, the power supply voltage output end Vout of the voltage reduction unit U5 outputs a power supply 1V1, a second end of the capacitor C7 and a second end of the capacitor C8 are connected to a power supply ground, respectively, and a ground terminal GND of the voltage reduction unit U5 is connected to the power supply ground.
The invention also discloses an intelligent image recognition swimming timing method based on machine learning, which comprises the following steps:
s1, acquiring image data information as the image to be processed;
s2, preprocessing the image to be processed obtained in the step S1 to obtain a preprocessed image;
s3, the image features in the preprocessed image in step S2 are extracted, and the individual swimmer information is calculated.
In a preferred embodiment of the present invention, in step S2, the method for preprocessing the image to be processed obtained by the method comprises the following steps:
s21, judging whether the acquired image is an RGB image:
if the acquired image is an RGB image, step S22 is executed;
if the acquired image is not an RGB image, performing step S23;
s22, converting the RGB image into a Grey image, wherein the method for converting the RGB image into the Grey image is as follows:
Grey＝r×R+g×G+b×B，
wherein, Grey represents a Grey image;
r represents a red component in an RGB image;
g represents a green component in an RGB image;
b represents a blue component in the RGB image;
r represents a red component coefficient;
g represents a green component coefficient;
b represents a blue component coefficient; r + g + b ═ 1; step S23 is executed;
s23, carrying out secondary processing on the image to obtain a preprocessed image; the secondary treatment method comprises the following steps:
s231, acquiring the size of an acquired image of Q multiplied by P, wherein Q is the width of the acquired image, and P is the height of the acquired image;
s232, the size of the intercepted image is qxp, wherein q is the width of the intercepted image, and p is the height of the intercepted image; q is a positive integer less than or equal to Q, and P is a positive integer less than or equal to P;
s233, calculating whether the number of the intercepted images is greater than or equal to a preset intercepted image number threshold value:
if the number of the intercepted images is greater than or equal to a preset intercepted image number threshold value, the size of the intercepted images meets the condition; the calculation method for the number of the intercepted images comprises the following steps:
Figure BDA0002924923610000051
zeta represents the number of the intercepted images;
int () represents a rounding function; step S234 is executed;
if the number of the intercepted images is smaller than a preset intercepted image number threshold value, the size of the images is intercepted again;
s234, sequencing the intercepted images in sequence as follows: 1 st cut imageI12 nd intercepting image I23 rd captured image I3… …, ζ th intercepted image Iζ；
Intercept image I for its xiξThe pixel values in the sequence are sorted from small to large, the xi is a positive integer less than or equal to zeta, and the xi is the 1 st pixel value
Figure BDA0002924923610000052
2 nd pixel value
Figure BDA0002924923610000053
Value of No. 3 pixel
Figure BDA0002924923610000054
… …, the first
Figure BDA0002924923610000056
Pixel value
Figure BDA0002924923610000055
S235, if
Figure BDA0002924923610000068
ψ ═ Z, Z representing a positive integer; then
Figure BDA0002924923610000061
Wherein the content of the first and second substances,
Figure BDA0002924923610000062
is shown as
Figure BDA0002924923610000069
Pixel value
Figure BDA0002924923610000063
If it is
Figure BDA00029249236100000610
Then
Figure BDA0002924923610000064
Wherein the content of the first and second substances,
Figure BDA0002924923610000065
is shown as
Figure BDA00029249236100000611
Pixel value
Figure BDA0002924923610000066
Is shown as
Figure BDA00029249236100000612
Pixel value
Figure BDA0002924923610000067
And S236, splicing the intercepted image obtained in the step S235 into a preprocessed image.
In a preferred embodiment of the present invention, in step S3, the underwater swimming time is calculated by:
s31, extracting the human body limb image feature points in the previous frame image of the frame image to be processed underwater or/and on the water surface;
s32, extracting the difference between the human body four-limb image feature point in the frame image to be processed and the human body four-limb feature point in the previous frame image according to the pixel point in the frame image to be processed and a preset difference threshold, wherein the difference between the human body four-limb image feature point in the frame image to be processed and the human body four-limb feature point in the previous frame image is the difference between the human body four-limb image feature point coordinate value in the frame image to be processed and the human body four-limb feature point coordinate value in the previous frame image, the preset difference threshold is obtained for the human body four-limb image feature point of the continuous epsilon frame images, epsilon is a positive integer which is more than or equal to 2 and is used for expressing the relationship between the pixel point in the image and the human body four-limb image feature point difference;
s33, obtaining the human body four limbs image feature points of the frame image to be processed according to the difference between the human body four limbs image feature points of the previous frame image and the human body four limbs image feature points;
s34, judging the swimmer to which the human body limb image characteristic point belongs according to the human body limb image characteristic point of the frame image to be processed obtained in the step S33;
and S35, obtaining the swimming time of the swimmer.
In a preferred embodiment of the present invention, in step S32, the preset difference threshold is calculated by:
Figure BDA0002924923610000071
wherein eta represents the initial difference of the image feature points of the four limbs of the human body;
epsilon represents the total number of continuous frame images;
Oj,0representing the human body limb image characteristic points of the jth continuous frame image;
Oj,1representing the human body limb image characteristic points of the original image corresponding to the jth continuous frame image;
αjrepresenting the weight value corresponding to the jth continuous frame image;
or/and in step S35, the swimming time is calculated by the following method:
Figure BDA0002924923610000072
wherein T represents swimming time;
u represents the total number of times of shooting;
Tuthe time when the swimmer enters water or touches the wall is shot at the current time;
Tu-1the time when the swimmer enters the water or touches the wall is photographed at the previous time.
In summary, due to the adoption of the technical scheme, the swimmer in the swimming lane can be timed according to the camera.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic block diagram of the connection of the present invention.
Fig. 2 is a schematic structural diagram of the present invention.
Fig. 3 is a schematic circuit diagram of a power module according to the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
The invention provides an intelligent image recognition swimming timing system based on machine learning, which comprises a swimming pool, wherein K lanes are arranged in the swimming pool, the K is a positive integer larger than or equal to 1 and is respectively a 1 st lane, a 2 nd lane, a 3 rd lane, … … th lane and a K th lane, preferably, the K is 10, namely 10 lanes are arranged in the swimming pool and are respectively a 1 st lane, a 2 nd lane, a 3 rd lane, a … … th lane and a 10 th lane, and an image acquisition display and a timer connected with the image acquisition display are arranged at the starting end or/and the opposite bank end of each lane; preferably, an image acquisition display and a timer connected with the image acquisition display are arranged at the starting end and the opposite end of each lane. In fig. 1, from left to right, the first image capturing display, the second image capturing display, the third image capturing display, the fourth image capturing display, the fifth image capturing display, the sixth image capturing display, the 8 th image capturing display, the 9 th image capturing display, the 10 th image capturing display, the 11 th image capturing display, the 12 th image capturing display, the 13 th image capturing display, the 14 th image capturing display, the 15 th image capturing display, the 16 th image capturing display, the 17 th image capturing display, the 18 th image capturing display, the 19 th image capturing display and the 20 th image capturing display are sequentially included from top to bottom; the 1 st image acquisition display is arranged at the initial end of the 1 st swimming lane, the 11 th image acquisition display is arranged at the opposite bank end of the 1 st swimming lane, the 2 nd image acquisition display is arranged at the initial end of the 2 nd swimming lane, the 12 th image acquisition display is arranged at the opposite bank end of the 2 nd swimming lane, the 3 rd image acquisition display is arranged at the initial end of the 3 rd swimming lane, the 13 th image acquisition display is arranged at the opposite bank end of the 3 rd swimming lane, the 4 th image acquisition display is arranged at the initial end of the 4 th swimming lane, the 14 th image acquisition display is arranged at the opposite bank end of the 4 th swimming lane, the 5 th image acquisition display is arranged at the initial end of the 5 th swimming lane, the 15 th image acquisition display is arranged at the opposite bank end of the 5 th swimming lane, the 6 th image acquisition display is arranged at the initial end of the 6 th swimming lane, the 16 th image acquisition display is arranged at the opposite bank end of the 6 th swimming lane, and the 7 th image acquisition, the 17 th image acquisition display is arranged at the opposite bank end of the 7 th lane, the 8 th image acquisition display is arranged at the starting end of the 8 th lane, the 18 th image acquisition display is arranged at the opposite bank end of the 8 th lane, the 9 th image acquisition display is arranged at the starting end of the 9 th lane, the 19 th image acquisition display is arranged at the opposite bank end of the 9 th lane, the 10 th image acquisition display is arranged at the starting end of the 10 th lane, and the 20 th image acquisition display is arranged at the opposite bank end of the 10 th lane;
the timing host, the score processing host and the large screen display are installed according to actual conditions, the timing host is respectively connected with the score processing host and each timer, and the score processing host is connected with the large screen display; and displaying the personal information and the corresponding achievement of the swimmer in each lane on a large-screen display. The personal information of the swimmer comprises swimming time information, lane information, names, ranks and belonged units. Swimming timing is mainly carried out through swimming time information;
in a preferred embodiment of the invention, the swimming pool further comprises a printer, the printer is connected with the achievement processing host computer in a wired or wireless mode, and the achievement processing host computer sends the personal information of the swimmers in each lane and the corresponding achievement to the printer for printing;
or/and the swimming pool further comprises a mobile terminal, wherein the mobile terminal is connected with the achievement processing host in a wireless mode, and the mobile terminal is used for checking the personal information and the corresponding achievement of each swimmer in each lane.
In a preferred embodiment of the invention, the device further comprises a starter and a sound box, and the sound box is used for emitting the starting sound of the starter. Its audio amplifier corresponds the number and is 10, be 1 audio amplifier respectively, the 2 nd audio amplifier, the 3 rd audio amplifier, the 4 th audio amplifier, the 5 th audio amplifier, the 6 th audio amplifier, the 7 th audio amplifier, the 8 th audio amplifier, 9 th audio amplifier and 10 th audio amplifier, the initiating terminal in the 1 st Lane is installed to its 1 st audio amplifier, the initiating terminal in the 2 nd Lane is installed to the 2 nd audio amplifier, the initiating terminal in the 3 rd Lane is installed to the 3 rd audio amplifier, the initiating terminal in the 4 th Lane is installed to the 4 th audio amplifier, the initiating terminal in the 5 th Lane is installed to the 5 th audio amplifier, the initiating terminal in the 6 th Lane is installed to the 6 th audio amplifier, the initiating terminal in the 7 th Lane is installed to the 7 th audio amplifier, the initiating terminal in the 8 th Lane is installed to the 8 th audio amplifier.
In a preferred embodiment of the present invention, as shown in fig. 2, the image capturing display includes a column disposed on the ground of the bank of the lane for supporting the panel, the top of the column is provided with the panel, the front of the panel is provided with a display screen mounting region for fixedly mounting the display screen, a speaker mounting region for fixedly mounting the speaker, a camera mounting region for mounting the camera, a button mounting region for mounting the button, and a numeric keypad mounting region for mounting the numeric keypad; the display screen is fixedly installed in the display screen installation area, the sound box is fixedly installed in the sound box installation area, the camera is fixedly installed in the camera installation area, the button is fixedly installed in the button installation area, and the numeric keyboard is installed in the numeric keyboard installation area;
a mainboard mounting seat for fixedly mounting a mainboard is arranged in the panel, the mainboard is fixedly mounted on the mainboard mounting seat, and a controller, a timer and a power supply module are arranged on the mainboard; the timing end of the controller is connected with the timing end of the timer, the display data end of the controller is connected with the display data end of the display screen, the sound box data end of the controller is connected with the data sound box end of the sound box, the image data input end of the controller is connected with the image data output end of the camera, the button end of the controller is connected with the data trigger end of the button, and the digital keyboard data end of the controller is connected with the data trigger end of the digital keyboard;
the underwater vehicle driving system also comprises a digital display screen, a water surface camera and an underwater camera which are arranged on the side walls of the lanes on the water surface and under the water, wherein the display data end of the digital display screen is connected with the digital display data end of the controller, and the image data output end of the underwater camera is connected with the underwater image data input end of the controller; the power supply module is respectively connected with a power supply end of the display screen, a power supply end of the sound box, a power supply end of the camera, a power supply end of the button, a power supply end of the numeric keyboard, a power supply end of the digital display screen and a power supply end of the underwater camera, and respectively supplies power to the display screen, the sound box, the camera, the button, the numeric keyboard, the digital display screen and the underwater camera;
and timing swimmers in the swim lane according to the camera.
In a preferred embodiment of the present invention, as shown in fig. 3, the power supply module includes: a power supply terminal VCC of the USB interface U1 is respectively connected with a first terminal of a resistor R1, a first terminal of a resistor R2, a first terminal of a resistor R4, an anode of a diode D1, a gate of a field effect transistor Q1 and a power supply terminal VCC of a power management unit U2, the power supply terminal of the USB interface U1 outputs a power supply V _ USB, a ground terminal GND of the USB interface U1 is connected with a power ground, a data positive terminal D +/DP of the USB interface U1 is connected with a USB data positive terminal of the base controller, and a data negative terminal D-/DM of the USB interface U1 is connected with a USB data negative terminal of the base controller; a second end of the resistor R4 is connected with the power ground;
a charging terminal CHRG of the power management unit U2 is respectively connected with a second terminal of the resistor R1 and a negative terminal of the charging indicator light LED1, a positive terminal of the charging indicator light LED1 is connected with a second terminal of the resistor R2, a battery terminal BAT of the power management unit U2 is respectively connected with a positive terminal of the battery holder U3 and a source of the field-effect transistor Q1, a negative terminal of the battery holder U3 is connected with a power ground, a current terminal PROG of the power management unit U2 is connected with a first terminal of the resistor R3, and a second terminal of the resistor R3 is respectively connected with the power ground and a ground terminal GND of the power management unit U2;
a drain electrode of the field effect transistor Q1 and a cathode electrode of the diode D1 are respectively connected with a first end of a capacitor C1, a first end of a capacitor C2, a first end of a resistor R5 and a power supply voltage input end Vin of the voltage reduction unit U4, a drain electrode of the field effect transistor Q1 and a cathode electrode of the diode D1 output a power supply VCC, a second end of the capacitor C1 and a second end of a capacitor C2 are respectively connected with a power supply ground, a second end of the resistor R5 is respectively connected with a first end of a resistor R6 and a first end of the capacitor C3, and a second end of the resistor R6 and a second end of the capacitor C3 are respectively connected with the power supply ground;
a power supply voltage output end Vout of the voltage reduction unit U4 is respectively connected with a first end of a capacitor C4, a first end of a capacitor C5, a first end of a capacitor C6 and a power supply voltage input end Vin of a voltage reduction unit U5, the power supply voltage output end Vout of the voltage reduction unit U4 outputs a power supply 3V3, a ground end GND of the voltage reduction unit U5 is connected with a power ground, and a second end of the capacitor C4, a second end of a capacitor C5 and a second end of the capacitor C6 are respectively connected with the power ground;
a power supply voltage output end Vout of the voltage reduction unit U5 is connected to a first end of the capacitor C7 and a first end of the capacitor C8, respectively, the power supply voltage output end Vout of the voltage reduction unit U5 outputs a power supply 1V1, a second end of the capacitor C7 and a second end of the capacitor C8 are connected to a power supply ground, respectively, and a ground terminal GND of the voltage reduction unit U5 is connected to the power supply ground.
The invention also discloses an intelligent image recognition swimming timing method based on machine learning, which comprises the following steps:
s1, acquiring image data information as the image to be processed;
s2, preprocessing the image to be processed obtained in the step S1 to obtain a preprocessed image;
s3, the image features in the preprocessed image in step S2 are extracted, and the individual swimmer information is calculated.
In a preferred embodiment of the present invention, in step S2, the method for preprocessing the image to be processed obtained by the method comprises the following steps:
s21, judging whether the acquired image is an RGB image:
if the acquired image is an RGB image, step S22 is executed;
if the acquired image is not an RGB image, performing step S23;
s22, converting the RGB image into a Grey image, wherein the method for converting the RGB image into the Grey image is as follows:
Grey＝r×R+g×G+b×B，
wherein, Grey represents a Grey image;
r represents a red component in an RGB image;
g represents a green component in an RGB image;
b represents a blue component in the RGB image;
r represents a red component coefficient;
g represents a green component coefficient;
b represents a blue component coefficient; r + g + b ═ 1; step S23 is executed;
s23, carrying out secondary processing on the image to obtain a preprocessed image; the secondary treatment method comprises the following steps:
s231, acquiring the size of an acquired image of Q multiplied by P, wherein Q is the width of the acquired image, and P is the height of the acquired image;
s232, the size of the intercepted image is qxp, wherein q is the width of the intercepted image, and p is the height of the intercepted image; q is a positive integer less than or equal to Q, and P is a positive integer less than or equal to P;
s233, calculating whether the number of the intercepted images is greater than or equal to a preset intercepted image number threshold value:
if the number of the intercepted images is greater than or equal to a preset intercepted image number threshold value, the size of the intercepted images meets the condition; the calculation method for the number of the intercepted images comprises the following steps:
Figure BDA0002924923610000131
zeta represents the number of the intercepted images;
int () represents a rounding function; step S234 is executed;
if the number of the intercepted images is smaller than a preset intercepted image number threshold value, the size of the images is intercepted again;
s234, sequencing the intercepted images in sequence as follows: 1 st cut image I12 nd intercepting image I23 rd captured image I3… …, ζ th intercepted image Iζ；
Intercept image I for its xiξThe pixel values in the sequence are sorted from small to large, the xi is a positive integer less than or equal to zeta, and the xi is the 1 st pixel value
Figure BDA0002924923610000132
2 nd pixel value
Figure BDA0002924923610000133
Value of No. 3 pixel
Figure BDA0002924923610000134
… …, the first
Figure BDA00029249236100001313
Pixel value
Figure BDA0002924923610000135
Figure BDA00029249236100001314
S235, if
Figure BDA00029249236100001315
ψ ═ Z, Z representing a positive integer; then
Figure BDA0002924923610000136
Wherein the content of the first and second substances,
Figure BDA0002924923610000137
is shown as
Figure BDA00029249236100001316
Pixel value
Figure BDA0002924923610000138
If it is
Figure BDA00029249236100001317
Then
Figure BDA0002924923610000139
Wherein the content of the first and second substances,
Figure BDA00029249236100001310
is shown as
Figure BDA00029249236100001318
Pixel value
Figure BDA00029249236100001311
Is shown as
Figure BDA00029249236100001319
Pixel value
Figure BDA00029249236100001312
And S236, splicing the intercepted image obtained in the step S235 into a preprocessed image.
In a preferred embodiment of the present invention, in step S3, the underwater swimming time is calculated by:
s31, extracting the human body limb image feature points in the previous frame image of the frame image to be processed underwater or/and on the water surface;
s32, extracting the difference between the human body four-limb image feature point in the frame image to be processed and the human body four-limb feature point in the previous frame image according to the pixel point in the frame image to be processed and a preset difference threshold, wherein the difference between the human body four-limb image feature point in the frame image to be processed and the human body four-limb feature point in the previous frame image is the difference between the human body four-limb image feature point coordinate value in the frame image to be processed and the human body four-limb feature point coordinate value in the previous frame image, the preset difference threshold is obtained for the human body four-limb image feature point of the continuous epsilon frame images, epsilon is a positive integer which is more than or equal to 2 and is used for expressing the relationship between the pixel point in the image and the human body four-limb image feature point difference;
s33, obtaining the human body four limbs image feature points of the frame image to be processed according to the difference between the human body four limbs image feature points of the previous frame image and the human body four limbs image feature points;
s34, judging the swimmer to which the human body limb image characteristic point belongs according to the human body limb image characteristic point of the frame image to be processed obtained in the step S33;
and S35, obtaining the swimming time of the swimmer.
In a preferred embodiment of the present invention, in step S32, the preset difference threshold is calculated by:
Figure BDA0002924923610000141
wherein eta represents the initial difference of the image feature points of the four limbs of the human body;
epsilon represents the total number of continuous frame images;
Oj,0representing the human body limb image characteristic points of the jth continuous frame image;
Oj,1representing the human body limb image characteristic points of the original image corresponding to the jth continuous frame image;
αjrepresenting the weight value corresponding to the jth continuous frame image;
or/and in step S35, the swimming time is calculated by the following method:
Figure BDA0002924923610000142
wherein T represents swimming time;
u represents the total number of times of shooting;
Tuthe time when the swimmer enters water or touches the wall is shot at the current time;
Tu-1the time when the swimmer enters the water or touches the wall is photographed at the previous time.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims (10)

1. An intelligent image recognition swimming timing system based on machine learning comprises a swimming pool, wherein K lanes are arranged in the swimming pool, are positive integers larger than or equal to 1 and are respectively a 1 st lane, a 2 nd lane, a 3 rd lane, … … and a K th lane, and the intelligent image recognition swimming timing system is characterized in that an image acquisition display and a timer connected with the image acquisition display are arranged at the starting end or/and the opposite end of each lane;
the system comprises a score processing host, a large screen, a plurality of timers, a timing host, a score processing host and a display, wherein the score processing host is connected with the large screen; and displaying the personal information and the corresponding achievement of the swimmer in each lane on a large screen.
2. The intelligent image recognition swimming timing system based on machine learning of claim 1, further comprising a printer, wherein the achievement processing host sends the personal information of the swimmers in each lane and the corresponding achievement to the printer for printing;
or/and the swimmer further comprises a mobile terminal, and the mobile terminal is used for checking the personal information and the corresponding achievement of each swimmer in each lane.
3. The machine learning based intelligent image recognition swimming timing system of claim 1, further comprising a commander and a sound box, wherein the sound box emits the sound of the commander.
4. The intelligent image recognition swimming timing system based on machine learning of claim 1, wherein K is 10, that is, there are 10 lanes in the swimming pool, that is, the 1 st lane, the 2 nd lane, the 3 rd lane, … …, and the 10 th lane, and there are an image acquisition display and a timer connected to the image acquisition display at the beginning and the opposite end of each lane.
5. The intelligent image recognition swimming timing system based on machine learning of claim 1, wherein the image acquisition display comprises a column arranged on the ground beside the lane for supporting a panel, the top end of the column is provided with the panel, the front face of the panel is provided with a display screen mounting area for fixedly mounting the display screen, a sound box mounting area for fixedly mounting a sound box, a camera mounting area for mounting a camera, a button mounting area for mounting a button, and a numeric keyboard mounting area for mounting a numeric keyboard; the display screen is fixedly installed in the display screen installation area, the sound box is fixedly installed in the sound box installation area, the camera is fixedly installed in the camera installation area, the button is fixedly installed in the button installation area, and the numeric keyboard is installed in the numeric keyboard installation area;
a mainboard mounting seat for fixedly mounting a mainboard is arranged in the panel, the mainboard is fixedly mounted on the mainboard mounting seat, and a controller, a timer and a power supply module are arranged on the mainboard; the timing end of the controller is connected with the timing end of the timer, the display data end of the controller is connected with the display data end of the display screen, the sound box data end of the controller is connected with the data sound box end of the sound box, the image data input end of the controller is connected with the image data output end of the camera, the button end of the controller is connected with the data trigger end of the button, and the digital keyboard data end of the controller is connected with the data trigger end of the digital keyboard;
the underwater vehicle driving system also comprises a digital display screen, a water surface camera and an underwater camera which are arranged on the side walls of the lanes on the water surface and under the water, wherein the display data end of the digital display screen is connected with the digital display data end of the controller, and the image data output end of the underwater camera is connected with the image data input end of the controller; the power supply module is respectively connected with a power supply end of the display screen, a power supply end of the sound box, a power supply end of the camera, a power supply end of the button, a power supply end of the numeric keyboard, a power supply end of the digital display screen and a power supply end of the underwater camera, and respectively supplies power to the display screen, the sound box, the camera, the button, the numeric keyboard, the digital display screen and the underwater camera;
and timing swimmers in the swim lane according to the camera.
6. The machine learning based intelligent image recognition swimming timing system of claim 1, wherein the power module comprises: a power supply terminal VCC of the USB interface U1 is respectively connected with a first terminal of a resistor R1, a first terminal of a resistor R2, a first terminal of a resistor R4, an anode of a diode D1, a gate of a field effect transistor Q1 and a power supply terminal VCC of a power management unit U2, the power supply terminal of the USB interface U1 outputs a power supply V _ USB, a ground terminal GND of the USB interface U1 is connected with a power ground, a data positive terminal D +/DP of the USB interface U1 is connected with a USB data positive terminal of the base controller, and a data negative terminal D-/DM of the USB interface U1 is connected with a USB data negative terminal of the base controller; a second end of the resistor R4 is connected with the power ground;
a charging terminal CHRG of the power management unit U2 is respectively connected with a second terminal of the resistor R1 and a negative terminal of the charging indicator light LED1, a positive terminal of the charging indicator light LED1 is connected with a second terminal of the resistor R2, a battery terminal BAT of the power management unit U2 is respectively connected with a positive terminal of the battery holder U3 and a source of the field-effect transistor Q1, a negative terminal of the battery holder U3 is connected with a power ground, a current terminal PROG of the power management unit U2 is connected with a first terminal of the resistor R3, and a second terminal of the resistor R3 is respectively connected with the power ground and a ground terminal GND of the power management unit U2;
a drain electrode of the field effect transistor Q1 and a cathode electrode of the diode D1 are respectively connected with a first end of a capacitor C1, a first end of a capacitor C2, a first end of a resistor R5 and a power supply voltage input end Vin of the voltage reduction unit U4, a drain electrode of the field effect transistor Q1 and a cathode electrode of the diode D1 output a power supply VCC, a second end of the capacitor C1 and a second end of a capacitor C2 are respectively connected with a power supply ground, a second end of the resistor R5 is respectively connected with a first end of a resistor R6 and a first end of the capacitor C3, and a second end of the resistor R6 and a second end of the capacitor C3 are respectively connected with the power supply ground;
a power supply voltage output end Vout of the voltage reduction unit U4 is respectively connected with a first end of a capacitor C4, a first end of a capacitor C5, a first end of a capacitor C6 and a power supply voltage input end Vin of a voltage reduction unit U5, the power supply voltage output end Vout of the voltage reduction unit U4 outputs a power supply 3V3, a ground end GND of the voltage reduction unit U5 is connected with a power ground, and a second end of the capacitor C4, a second end of a capacitor C5 and a second end of the capacitor C6 are respectively connected with the power ground;
a power supply voltage output end Vout of the voltage reduction unit U5 is connected to a first end of the capacitor C7 and a first end of the capacitor C8, respectively, the power supply voltage output end Vout of the voltage reduction unit U5 outputs a power supply 1V1, a second end of the capacitor C7 and a second end of the capacitor C8 are connected to a power supply ground, respectively, and a ground terminal GND of the voltage reduction unit U5 is connected to the power supply ground.
7. An intelligent image recognition swimming timing method based on machine learning is characterized by comprising the following steps:
s1, acquiring image data information as the image to be processed;
s2, preprocessing the image to be processed obtained in the step S1 to obtain a preprocessed image;
s3, the image features in the preprocessed image in step S2 are extracted, and the time for the swimmer to enter the water and touch the wall is calculated.
8. The intelligent image recognition swimming timing method based on machine learning of claim 7, wherein in step S2, the method for preprocessing the image to be processed obtained by the method comprises the following steps:
s21, judging whether the acquired image is an RGB image:
if the acquired image is an RGB image, step S22 is executed;
if the acquired image is not an RGB image, performing step S23;
s22, converting the RGB image into a Grey image, wherein the method for converting the RGB image into the Grey image is as follows:
Grey＝r×R+g×G+b×B，
wherein, Grey represents a Grey image;
r represents a red component in an RGB image;
g represents a green component in an RGB image;
b represents a blue component in the RGB image;
r represents a red component coefficient;
g represents a green component coefficient;
b represents a blue component coefficient; r + g + b ═ 1; step S23 is executed;
s23, carrying out secondary processing on the image to obtain a preprocessed image; the secondary treatment method comprises the following steps:
s231, acquiring the size of an acquired image of Q multiplied by P, wherein Q is the width of the acquired image, and P is the height of the acquired image;
s232, the size of the intercepted image is qxp, wherein q is the width of the intercepted image, and p is the height of the intercepted image; q is a positive integer less than or equal to Q, and P is a positive integer less than or equal to P;
s233, calculating whether the number of the intercepted images is greater than or equal to a preset intercepted image number threshold value:
if the number of the intercepted images is greater than or equal to a preset intercepted image number threshold value, the size of the intercepted images meets the condition; the calculation method for the number of the intercepted images comprises the following steps:
Figure FDA0002924923600000041
zeta represents the number of the intercepted images;
int () represents a rounding function; step S234 is executed;
if the number of the intercepted images is smaller than a preset intercepted image number threshold value, the size of the images is intercepted again;
s234, sequencing the intercepted images in sequence as follows: 1 st cut image I12 nd intercepting image I23 rd captured image I3… …, ζ th intercepted image Iζ；
To its sixth placeIntercepting an image IξThe pixel values in the sequence are sorted from small to large, the xi is a positive integer less than or equal to zeta, and the xi is the 1 st pixel value
Figure FDA0002924923600000051
2 nd pixel value
Figure FDA0002924923600000052
Value of No. 3 pixel
Figure FDA0002924923600000053
First, the
Figure FDA0002924923600000054
Pixel value
Figure FDA0002924923600000055
S235, if
Figure FDA0002924923600000056
ψ ═ Z, Z representing a positive integer; then
Figure FDA0002924923600000057
Wherein the content of the first and second substances,
Figure FDA0002924923600000058
is shown as
Figure FDA0002924923600000059
Pixel value
Figure FDA00029249236000000510
If it is
Figure FDA00029249236000000511
Then
Figure FDA00029249236000000512
Wherein the content of the first and second substances,
Figure FDA00029249236000000513
is shown as
Figure FDA00029249236000000514
Pixel value
Figure FDA00029249236000000515
Is shown as
Figure FDA00029249236000000516
Pixel value
Figure FDA00029249236000000517
And S236, splicing the intercepted image obtained in the step S235 into a preprocessed image.
9. The intelligent image recognition swimming timing method based on machine learning of claim 7, wherein in step S3, the underwater swimming time is calculated by:
s31, extracting the human body limb image feature points in the previous frame image of the frame image to be processed underwater or/and on the water surface;
s32, extracting the difference between the human body four-limb image feature point in the frame image to be processed and the human body four-limb feature point in the previous frame image according to the pixel point in the frame image to be processed and a preset difference threshold, wherein the difference between the human body four-limb image feature point in the frame image to be processed and the human body four-limb feature point in the previous frame image is the difference between the human body four-limb image feature point coordinate value in the frame image to be processed and the human body four-limb feature point coordinate value in the previous frame image, the preset difference threshold is obtained for the human body four-limb image feature point of the continuous epsilon frame images, epsilon is a positive integer which is more than or equal to 2 and is used for expressing the relationship between the pixel point in the image and the human body four-limb image feature point difference;
s33, obtaining the human body four limbs image feature points of the frame image to be processed according to the difference between the human body four limbs image feature points of the previous frame image and the human body four limbs image feature points;
s34, judging the swimmer to which the human body limb image characteristic point belongs according to the human body limb image characteristic point of the frame image to be processed obtained in the step S33;
and S35, obtaining the swimming time of the swimmer.
10. The intelligent image recognition swimming timing method based on machine learning of claim 7, wherein in step S32, the preset difference threshold is calculated by:
Figure FDA0002924923600000061
wherein eta represents the initial difference of the image feature points of the four limbs of the human body;
epsilon represents the total number of continuous frame images;
Oj,0representing the human body limb image characteristic points of the jth continuous frame image;
Oj,1representing the human body limb image characteristic points of the original image corresponding to the jth continuous frame image;
αjrepresenting the weight value corresponding to the jth continuous frame image;
or/and in step S35, the swimming time is calculated by the following method:
Figure FDA0002924923600000062
wherein T represents swimming time;
u represents the total number of water entries or wall contacts;
Tuthe time when the swimmer enters water or touches the wall is shot at the current time;
Tu-1the time when the swimmer enters the water or touches the wall is photographed at the previous time.
CN202110129228.0A 2021-01-29 2021-01-29 Intelligent image recognition swimming timing system and timing method based on machine learning Active CN112818842B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110129228.0A CN112818842B (en) 2021-01-29 2021-01-29 Intelligent image recognition swimming timing system and timing method based on machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110129228.0A CN112818842B (en) 2021-01-29 2021-01-29 Intelligent image recognition swimming timing system and timing method based on machine learning

Publications (2)

Publication Number Publication Date
CN112818842A true CN112818842A (en) 2021-05-18
CN112818842B CN112818842B (en) 2024-06-14

Family

ID=75860335

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110129228.0A Active CN112818842B (en) 2021-01-29 2021-01-29 Intelligent image recognition swimming timing system and timing method based on machine learning

Country Status (1)

Country Link
CN (1) CN112818842B (en)

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07271986A (en) * 1994-03-28 1995-10-20 Yusaku Fujii Pool monitoring method
JPH1166319A (en) * 1997-08-21 1999-03-09 Omron Corp Method and device for detecting traveling object, method and device for recognizing traveling object, and method and device for detecting person
JP2000050245A (en) * 1998-03-31 2000-02-18 Seiko Instruments Inc Underwater photographing device, and underwater photographing and recording system for swimming race using the underwater photographing device
JP2002277481A (en) * 2001-03-15 2002-09-25 Hamamatsu Photonics Kk Traveling measuring instrument
JP2005331438A (en) * 2004-05-21 2005-12-02 Seiko Precision Inc Measuring system for competitive sport and measurement method
CN105806320A (en) * 2014-12-29 2016-07-27 同方威视技术股份有限公司 Shooting measure system and method
CN108345821A (en) * 2017-01-24 2018-07-31 成都理想境界科技有限公司 Face tracking method and apparatus
CN110084204A (en) * 2019-04-29 2019-08-02 北京字节跳动网络技术有限公司 Image processing method, device and electronic equipment based on target object posture
CN110180151A (en) * 2019-05-06 2019-08-30 南昌嘉研科技有限公司 A kind of swimming instruction auxiliary system
CN110778169A (en) * 2019-11-06 2020-02-11 李洪林 Swimming pool starting platform, swimming training system and training method
CN111753749A (en) * 2020-06-28 2020-10-09 华东师范大学 Lane line detection method based on feature matching
CN214586929U (en) * 2021-01-29 2021-11-02 徐文海 Intelligent image recognition swimming timing system

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07271986A (en) * 1994-03-28 1995-10-20 Yusaku Fujii Pool monitoring method
JPH1166319A (en) * 1997-08-21 1999-03-09 Omron Corp Method and device for detecting traveling object, method and device for recognizing traveling object, and method and device for detecting person
JP2000050245A (en) * 1998-03-31 2000-02-18 Seiko Instruments Inc Underwater photographing device, and underwater photographing and recording system for swimming race using the underwater photographing device
JP2002277481A (en) * 2001-03-15 2002-09-25 Hamamatsu Photonics Kk Traveling measuring instrument
JP2005331438A (en) * 2004-05-21 2005-12-02 Seiko Precision Inc Measuring system for competitive sport and measurement method
CN105806320A (en) * 2014-12-29 2016-07-27 同方威视技术股份有限公司 Shooting measure system and method
CN108345821A (en) * 2017-01-24 2018-07-31 成都理想境界科技有限公司 Face tracking method and apparatus
CN110084204A (en) * 2019-04-29 2019-08-02 北京字节跳动网络技术有限公司 Image processing method, device and electronic equipment based on target object posture
CN110180151A (en) * 2019-05-06 2019-08-30 南昌嘉研科技有限公司 A kind of swimming instruction auxiliary system
CN110778169A (en) * 2019-11-06 2020-02-11 李洪林 Swimming pool starting platform, swimming training system and training method
CN111753749A (en) * 2020-06-28 2020-10-09 华东师范大学 Lane line detection method based on feature matching
CN214586929U (en) * 2021-01-29 2021-11-02 徐文海 Intelligent image recognition swimming timing system

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
ADMIN: "游泳计时记分系统", pages 1 - 8, Retrieved from the Internet <URL:http://www.sportsoft.cn/11666.html> *
JIANBIN HOU等: "RETRACTED：Swimming target detection and tracking technology in video image processing", 《MICROPROCESSORS AND MICROSYSTEMS》, 2 March 2021 (2021-03-02), pages 1 - 10 *
KARIN DE LANGIS等: "Realtime multi-diver tracking and re-identification for underwater human robot collaboration", 《2020 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION（ICRA）》, 15 September 2020 (2020-09-15), pages 11140 - 11146 *
关静等: "基于角点检测的规则几何图形识别算法研究", 《现代信息科技》, vol. 3, no. 24, 25 December 2019 (2019-12-25), pages 71 - 73 *
神州数码: "半自动计时系统在游泳比赛中的应用", 《游泳》, no. 04, 20 July 2006 (2006-07-20), pages 51 - 52 *

Also Published As

Publication number Publication date
CN112818842B (en) 2024-06-14

Similar Documents

Publication Publication Date Title
CN108734104B (en) Body-building action error correction method and system based on deep learning image recognition
CN111931701B (en) Gesture recognition method and device based on artificial intelligence, terminal and storage medium
CN109274883A (en) Posture antidote, device, terminal and storage medium
CN105631417A (en) Video beautification system and method applied to Internet video live broadcast
CN112364842B (en) Double-shot face recognition method and device
EP1768063B8 (en) Image checking device, image checking method, and image checking program
CN214586929U (en) Intelligent image recognition swimming timing system
CN109559362B (en) Image subject face replacing method and device
CN109117753A (en) Position recognition methods, device, terminal and storage medium
CN109028235A (en) Intelligent Wind Speed adjustable type kitchen ventilator based on background modeling
CN108513074B (en) Self-photographing control method and device and electronic equipment
EP1471456A3 (en) Method and apparatus for finding optimal threshold for image segmentation
US20110246204A1 (en) Image display device and method thereof
CN112818842B (en) Intelligent image recognition swimming timing system and timing method based on machine learning
TW202209175A (en) Image correction method and system based on deep learning
WO2016107226A1 (en) Image processing method and apparatus
CN115147339A (en) Human body key point detection method and related device
EP4020389A1 (en) Target user locking method and electronic device
CN112700568A (en) Identity authentication method, equipment and computer readable storage medium
CN112439180B (en) Intelligent voice playing method and equipment
CN109087694B (en) Method for helping students to exercise bodies and family education equipment
EP1377021A3 (en) Image data processing method, portable terminal apparatus, and computer program
US8974296B2 (en) Game program and game apparatus
CN108615219A (en) A kind of image landscaping treatment method of intelligent terminal
CN112802051A (en) Fitting method and system of basketball shooting curve based on neural network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant