CPC G06F 18/214 (2023.01) [A01M 99/00 (2013.01); G06T 7/11 (2017.01); G06V 10/25 (2022.01); G06T 2207/20084 (2013.01); G06T 2207/20221 (2013.01); G06T 2207/30188 (2013.01); G06T 2207/30242 (2013.01); G06V 10/225 (2022.01); G06V 2201/03 (2022.01)] | 20 Claims |
1. An object counting method based on an image, comprising:
cutting the image into a plurality of split images;
inputting the split images into a neural network to identify an object from the split images;
merging a first split image of the split images and a second split image of the split images into a merged image in response to a distance between at least one side of a bounding box being associated with a position of a first object in the first split image and at least one side of a bounding box being associated with a position of a second object in the second split image adjacent to the first split image is less than a tolerance value;
merging the first object and the second object into a merged object and remarking a position of the merged object; and
calculating a prediction number of the object according to the object in the split images and the merged object in the merged image included in the split images;
wherein calculating the prediction number of the object comprises:
calculating the prediction number of the object by subtracting a number of the merged object in the merged image from a number of the object in the split images.
|