Methods, systems, and apparatuses, including computer programs encoded on a computer-readable storage medium for estimating the shape, size, mass, and health of fish are described. A pair of stereo cameras may be utilized to obtain off-axis images of fish in a defined area. The images may be processed, enhanced, and combined. Object detection may be used to detect and track a fish in images. A pose estimator may be used to determine key points and features of the detected fish. Based on the key points, a model of the fish is generated that provides an estimate of the size and shape of the fish. A regression model or neural network model can be applied to the fish model to determine characteristics of the fish.

The size estimation device 5 is provided with an estimation unit 51 and a calculation unit 52. The estimation unit 51 estimates, on the basis of a learning model which learns the feature of a first underwater creature body that faces one direction in an image captured in the water, a first feature point that indicates one end in a one direction side of a second underwater creature body that faces a one direction included in an image captured outside and a second feature point that indicates the other end. The calculation unit 52 calculates information indicating the size of the second underwater creature body on the basis of information including the first feature point and the second feature point.

Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for sea lice mitigation. In some implementations, a method includes obtaining multiple observations of a population of reference fish across a period of time, generating, from the multiple observations, a record for each reference fish that indicates an extent of sea lice infestation for the reference fish across the period of time, training, based on the records, a model that determines a predicted health indicator for a fish, obtaining an image of a sample fish that is not in the population of reference fish, determining, based on the image of the sample fish and with the model, a predicted health indicator for the sample fish, and selectively initiating sea lice mitigation based on the predicted health indicator.

Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for sea lice mitigation. In some implementations, a method includes obtaining multiple observations of a population of reference fish across a period of time, generating, from the multiple observations, a record for each reference fish that indicates an extent of sea lice infestation for the reference fish across the period of time, training, based on the records, a model that determines a predicted health indicator for a fish, obtaining an image of a sample fish that is not in the population of reference fish, determining, based on the image of the sample fish and with the model, a predicted health indicator for the sample fish, and selectively initiating sea lice mitigation based on the predicted health indicator.

Methods and systems are disclosed for improvements in aquaculture that allow for increasing the number and growth efficiency of fish in an aquaculture setting (or other animals in other settings) by identifying or predicting characteristics of the animals based on visual or ultrasound images of the animals obtained through non-invasive means. The animals are sorted based on the characteristics.

Methods and systems are disclosed for improvements in aquaculture that allow for increasing the number and growth efficiency of fish in an aquaculture setting (or other animals in other settings) by identifying or predicting characteristics of the animals based on visual or ultrasound images of the animals obtained through non-invasive means. The animals are sorted based on the characteristics.

The present invention relates to a method and apparatus for providing a dynamic decision-making process in relation to feeding animals in water. More particularly, the present invention relates to a method and apparatus for improving feeding and/or farming strategies used in a fish farm. According to a first aspect, there is provided a computer-implemented method for feeding one or more aquatic animals, the method comprising the steps of: receiving pre-processed sensor data in relation to the one or more aquatic animals; inputting the pre-processed sensor data into one or more learned decision-making models, wherein the one or more learned decision-making models has been trained to substantially optimise the rate and amount of food provided to the aquatic animals; determining, by the one or more learned decision-making models using the received pre-processed sensor data, feeding instructions for the one or more aquatic animals; and outputting the feeding instructions from the one or more learned decision-making models.

The present invention relates to a method, a system and an apparatus for treating food items. Such method and device may, for instance, be used to cool down food items, bleed freshly slaughtered fish and treat live fish against parasites. The device and method of the present invention further uses temperature differences and agitation to tread food items. A method for reducing or eliminating sea lice from salmon fish, said method comprising: a) transporting the salmon fish from a reservoir and the feeding salmon fish in liquid into the tubing of a first pumping device, said liquid being at a temperature from +15 to +28° C., b) transporting the salmon fish in liquid through the tubing of the pumping device at a first set time period for treatment in said liquid, c) feeding the salmon fish out of the first pumping device, d) separating the liquid from the salmon fish, and e) transporting the salmon fish back to the reservoir, wherein the liquid media from the pumping device is redirected after separation from the salmon fish in step d) through a heat exchange element back into the pumping device for continued treatment of further salmon fish, and wherein the temperature difference between the liquid used in steps b) and the reservoir is more than 15° C.

The present invention relates to a method, a system and an apparatus for treating food items. Such method and device may, for instance, be used to cool down food items, bleed freshly slaughtered fish and treat live fish against parasites. The device and method of the present invention further uses temperature differences and agitation to tread food items. A method for reducing or eliminating sea lice from salmon fish, said method comprising: a) transporting the salmon fish from a reservoir and the feeding salmon fish in liquid into the tubing of a first pumping device, said liquid being at a temperature from +15 to +28° C., b) transporting the salmon fish in liquid through the tubing of the pumping device at a first set time period for treatment in said liquid, c) feeding the salmon fish out of the first pumping device, d) separating the liquid from the salmon fish, and e) transporting the salmon fish back to the reservoir, wherein the liquid media from the pumping device is redirected after separation from the salmon fish in step d) through a heat exchange element back into the pumping device for continued treatment of further salmon fish, and wherein the temperature difference between the liquid used in steps b) and the reservoir is more than 15° C.

Animal sensing systems and methods are described. An animal sensing system can include an input section with an entrance and an output section having at least two output paths each having its own exit. An animal can enter the animal sensing system through the entrance. A sensor within a sensing area of the input section can detect one or more characteristics of the animal, and can communicate the detected characteristic(s) to a central processing unit. The central processing unit can use the received data to classify the animal based on the detected characteristic(s), and then control a directional guide, such as a gate, in the animal sensing system on the basis of the classification, so as to direct the animal within the animal sensing system, such as to allow access to only one of the output paths at a time. The animal may thus be allowed to exit the animal sensing system through only one output path, directing the animal to a desired location based on the classification.