Browse Prior Art Database

Method for Providing Lap Counter with High Speed Cameras, Image Detectors and Classifier to Identify Objects and Timing Data

IP.com Disclosure Number: IPCOM000250360D
Publication Date: 2017-Jul-05
Document File: 2 page(s) / 165K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method is disclosed for providing lap counter with high speed cameras, image detectors and classifier to identify objects and timing data.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

1

Method for Providing Lap Counter with High Speed Cameras, Image Detectors and Classifier to Identify Objects and Timing Data

In general, lap counters used for counting laps of an object such as, but not limited to, person, dog and car is identified by an application installed in a portable communication device such as, but not limited, mobile camera and tablet camera. However, the application is unable to differentiate competitors or objects being tracked by the portable communication device. Further, video based timing system and video based scoring system has come into existence to monitor colors of unique objects and associate time stamps to the objects. However, the video based timing system and video based scoring system are not intelligent and are limited to use of specific colors and are also unable to identify unique objects from a video frame.

Disclosed is a method for providing lap counter with high speed cameras, image detectors and a classifier to identify objects and timing data. The lap counter can be a visual timing system which is positioned at a finish line or a start line of a race to recognize objects passing through the visual timing system for assigning the timing data. The visual timing system is also used to analyze the objects trajectory to predict future placement of the objects in the race. The time and placement of the objects is analyzed by prior training of the visual timing system to recognize the objects being tracked such as, but not limited to, airplanes, drones, racer, bib number, and car. Further, the visual timing system used at check points of a race course, predicts time for the next check point based on the movement of the object in the race. The predicted time is relayed to a racer to adjust course of the race or to estimate the race completion time.

Fig. 1 illustrates a multi-layer convolutional neural network used to build the classifier.

Fig. 1

As illustrated in fig. 1, the classifier includes multiple convolution and pooling layers with several neurons. Each neuron in the classifier is followed by two...