Imaging and three dimensional reconstruction for weight estimation

A weighing station device for estimating a weight of an animal, the device comprising:

• a stall for the animal;
• a camera in the stall that is activated to capture a plurality of digital images of the animal when a presence of the animal is detected at the stall;
• and
• a processor configured to:
• obtain the plurality of digital images of the animal from the camera;
• select a set of digital images from the plurality of digital images to construct a three-dimensional digital image based at least in part on a quality of the plurality of digital images;
• wherein selecting the set of digital images comprises:
• obtain digital images of the animal with the camera at a plurality of locations with respect to the animal;
• determine a level of obstruction at each of the plurality of locations;
• skip capture of the animal at a current location of the plurality of locations based at least in part on the determined level of obstruction;
• and
• construct the three-dimensional digital image of the animal from the set of digital images;
• determine a dimension of the animal from the three-dimensional digital image that is based at least in part on the set of digital images;
• and
• estimate the weight of the animal from the determined dimension and a model relating the dimension of the animal to the weight of the animal.

A computer program product for weighing an animal, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform:

• activating a camera to capture a plurality of digital images of the animal obtained at a plurality of locations with respect to the animal when the animal is detected at a stall for the animal;
• selecting a set of digital images from the plurality of digital images to construct a three-dimensional digital image based at least in part on a quality of the plurality of digital images,
• wherein selecting the set of digital images comprises:
• obtaining digital images of the animal with the camera at the plurality of locations with respect to the animal;
• determining a level of obstruction at each of the plurality of locations;
• skipping capture of the animal at a current location of the plurality of locations based at least in part on the determined level of obstruction;
• and
• constructing the three-dimensional digital image of the animal from the set of digital images;
• determining a dimension of the animal from the three-dimensional digital image that is based at least in part on the set of digital images;
• ranking the weight of all animals based on their dimensions;
• and
• estimating the weight of the animal from the determined dimension and a model relating the dimension of the animal to the weight of the animal.