Object detecting device, object detecting method and non-transitory computer-readable medium

An object detecting method, comprising:

• generating, by a classifier, a current color image and a current gray scale image;
• generating, by the classifier, an initial characteristic pattern from the current color image via a neural network algorithm;
• adjusting, by the classifier, a current dimension of the initial characteristic pattern to generate an adjusted characteristic pattern according to a gray scale image dimension of the current gray scale image;
• concatenating, by the classifier, the adjusted characteristic pattern and the current gray scale image to calculate a class confidence;
• determining, by the classifier, whether the class confidence is larger than a confidence threshold, and outputting, by the classifier, a current classification result if the class confidence is larger than the confidence threshold;
• and
• storing, by a storage device, the current classification result.

A non-transitory computer readable medium, having a plurality of program codes for performing an object detecting method, wherein the object detecting method comprises:

• generating, by a classifier, a current color image and a current gray scale image;
• generating, by the classifier, an initial characteristic pattern from the current color image via a neural network algorithm;
• adjusting, by the classifier, a current dimension of the initial characteristic pattern to generate an adjusted characteristic pattern according to a gray scale image dimension of the current gray scale image;
• concatenating, by a classifier, the adjusted characteristic pattern and the current gray scale image to calculate a class confidence;
• determining, by the classifier, whether the class confidence is larger than a confidence threshold, and outputting, by the classifier, a current classification result if the class confidence is larger than the confidence threshold;
• and
• storing, by a storage device, the current classification result.