Computer vision system and method for agriculture
There is provided a computer vision monitoring system and method for use in agriculture. The system has a frame; a battery mounted to the frame and providing electrical power; a Light Detection and Ranging (LiDAR) sensor mounted to the frame; a camera mounted to the frame; an enclosure mounted to the frame, the enclosure housing a processing structure. The processing structure executes instructions such as capturing point data from the LiDAR sensor; capturing image data from the camera; generating a 3D color-assigned cloud point data set from the image data and the point data; and determining crop characteristic data. The method involves similar steps as executed by the processing structure.
1. A crop monitoring system for a field comprising:
a frame;
a battery mounted to the frame and providing electrical power;
a Light Detection and Ranging (LiDAR) sensor mounted to the frame;
a camera mounted to the frame;
an enclosure mounted to the frame, the enclosure housing a processing structure; the processing structure executing a plurality of instructions stored on a tangible computer-readable memory;
the instructions comprise:
capturing point data from the LiDAR sensor;
capturing image data from the camera;
generating a 3D color-assigned cloud point data set from the image data and the point data; and
determining crop characteristic data based on the 3D color-assigned cloud point data set.
2. The crop monitoring system according to claim 1 , further comprising: a rotating platform mounted on the frame; the rotating platform supporting the LiDAR sensor and the camera; and the instructions further comprise: rotating the rotating platform before the capturing of the point data and the image data.
3. The crop monitoring system according to claim 2 , wherein the instructions further comprise: determining a field of view (FOV) of the LiDAR sensor; determining a rotational measurement associated with the FOV of the LiDAR sensor; and rotating the rotating platform by the rotational measurement before the capturing of the point data.
4. The crop monitoring system according to claim 1 , wherein the instructions further comprise: filtering the 3D color-assigned cloud point data set prior to determining the crop characteristic data.
5. The crop monitoring system according to claim 1 , wherein the instructions further comprise: filtering the point data set prior to generating the 3D color-assigned cloud point data set.
6. The crop monitoring system according to claim 1 , wherein the instructions further comprise: segmenting crop data points from the 3D color-assigned cloud point data set as part of the determining of the crop characteristic data.
7. The crop monitoring system according to claim 1 , wherein the processing structure comprises a graphical processing unit (GPU).
8. The crop monitoring system according to claim 1 , further comprising: a solar panel providing electrical charge to the battery.
9. The crop monitoring system according to claim 8 , wherein the instructions further comprise: a battery management system placing the processing structure in a standby state when the battery management system determines that the battery falls close to 80% of full capacity.
10. The crop monitoring system according to claim 1 , further comprising: at least one sensor providing at least one sensor measurement to the processing structure; and the instructions further comprise: determining at least one parameter to adjust the crop characteristic data.
11. The crop monitoring system according to claim 10 , wherein the at least one sensor is selected from at least one of: a temperature sensor, a humidity sensor, a pressure sensor, a soil moisture sensor, a soil resistance sensor, a nitrogen/phosphorus/potassium (NPK) sensor, a rain gauge, and a solar radiation sensor.
12. A method for monitoring a field comprises:
capturing point data from a Light Detection and Ranging (LiDAR) sensor;
capturing image data from a camera;
generating, using a processor, a 3D color-assigned cloud point data set from the image data and the point data; and
determining, using the processor, crop characteristic data based on the 3D color-assigned cloud point data set.
13. The method of claim 12 further comprise: rotating a rotating platform supporting the LiDAR sensor and the camera before the capturing of the point data and the image data.
14. The method of claim 13 further comprise: determining a field of view (FOV) of the LiDAR sensor; determining a rotational measurement associated with the FOV of the LiDAR sensor; and rotating the rotating platform by the rotational measurement before the capturing of the point data.
15. The method of claim 12 further comprise: filtering, by the processor, the 3D color-assigned cloud point data set prior to determining the crop characteristic data.
16. The method of claim 12 further comprise: filtering, by the processor, the point data set prior to generating the 3D color-assigned cloud point data set.
17. The method of claim 12 further comprise: segmenting, by the processor, crop data points from the 3D color-assigned cloud point data set as part of the determining of the crop characteristic data.
18. The method of claim 12 , wherein the processor comprises a graphical processing unit (GPU).
19. The method of claim 12 further comprise: placing the processor in a standby state when a battery management system determines that the battery falls close to 80% of full capacity.
20. The method of claim 12 , further comprise: reading at least one sensor measurement from at least one sensor; and determining at least one parameter from the at least one sensor measurement to adjust the crop characteristic data.
21. The method of claim 20 , wherein the at least one sensor is selected from at least one of: a temperature sensor, a humidity sensor, a pressure sensor, a soil moisture sensor, a soil resistance sensor, a nitrogen/phosphorus/potassium (NPK) sensor, a rain gauge, and a solar radiation sensor.