Method for recognizing failure detection of e.g. object in image, involves recognizing whether captured object is erroneously detected, where recognition is performed based on determined property of captured object

Abstract
(A1) Ein Verfahren zum Erkennen einer Fehldetektion eines Objekts in einem Bild umfasst ein Erfassen eines Objekts in einem Bild einer Mehrzahl von Bildern und ein Analysieren der Mehrzahl von Bildern, um eine Eigenschaft des erfassten Objekts zu bestimmen. Dabei wird fuer jedes Bild der Mehrzahl von Bildern bestimmt, ob das jeweilige Bild das erfasste Objekt zeigt. Des Weiteren umfasst das Verfahren ein Erkennen, dass es sich bei dem erfassten Objekt um eine Fehldetektion handelt, wobei das Erkennen auf der bestimmten Eigenschaft des erfassten Objekts basiert.

 

DE 102009048117 A1 UPAB: 20110425 NOVELTY - The method involves capturing an object in one of a set of images and analyzing the images to determine a property of the captured object. A determination is made whether the respective image shows the captured object for each of the images. Recognition is made whether the captured object is erroneously detected, where the recognition is performed based on the determined property of the captured object. Quality of the object is determined while analyzing the images, and failure detection is identified based on the determined quality of the object. DETAILED DESCRIPTION - INDEPENDENT CLAIMS are also included for the following: (1) an image recognition system comprising a device for recognizing failure detection of an object in an image (2) a computer program with a program code provided with a set of instructions for executing a method for recognizing failure detection of an object in an image. USE - Method for recognizing failure detection of an object in an image recorded with a single static camera or a mobile camera. Uses include but are not limited to a human, a face of the human, an animal and a vehicle. ADVANTAGE - The method enables performing the failure detection of the object in the images in a reliable and rapid manner with less effort, and enables increasing probability of performing the failure detection.