System and method for measuring leaf-to-stem ratio
Systems and methods are provided for determining a leaf-to-stem ratio of a crop automatically from an image of the crop. An agricultural vehicle may include a crop analysis system having a computing device and an imaging device for capturing an image of a crop. The computing device analyzes images acquired by the imaging determine to determine a leaf-to-stem ratio of the crop.
1 . A crop analysis system, comprising:
an imaging device coupled to an agricultural vehicle and configured to acquire an image of a crop during harvesting by the agricultural vehicle as the agricultural vehicle traverses a field including the crop, wherein the imaging device is a high-speed camera; and
a computing device having a processor executing computer-readable instructions to:
analyze the image of the crop to identify portions of the image corresponding to leaves and identify portions of the image corresponding to stems;
generate a first metric indicating an amount of leaf in the image of the crop;
generate a second metric indicating an amount of stem in the image of the crop;
determine a leaf-to-stem ratio for the crop based on the first and second metric; and
signal an agricultural machine to perform an action subsequent to harvesting based on the leaf-to-stem ratio, wherein the agricultural machine includes one of a baler.
2 . The crop analysis system of claim 1 , wherein the computing device is further configured to identify the portions of the image respectively corresponding to leaves and stem by detecting edges in the image base on morphological processing.
3 . The crop analysis system of claim 2 , wherein the computing device is further configured to apply a machine learning technique to distinguish leaves and stems following edge detection.
4 . The crop analysis system of claim 1 , wherein the computing device is further configured to identify the portions of the image respectively corresponding to leaves and stem through image segmentation.
5 . The crop analysis system of claim 1 , wherein the computing device is further configured to apply a plurality of markers on the image and generate a count of marker overlaying leaves and a count of markers overlaying stems.
6 . The crop analysis system of claim 5 , wherein the computing device is further configured to determine the leaf-to-stem ratio based on the respective counts of markers overlaying leaves and stems.
7 . The crop analysis system of claim 1 , wherein the computing device is further configured to apply an image area measurement corresponding to leaves and an image area measurement corresponding to stems.
8 . The crop analysis system of claim 7 , wherein the computing device is further configured to determine the leaf-to-stem ratio based on the respective image area measurements corresponding to leaves and stems.
9 . The crop analysis system of claim 1 , wherein the computing device is further configured to acquire a plurality of images via the imaging device and determine an aggregate leaf-to-stem ratio corresponding to the plurality of images.
10 . A crop analysis method, comprising:
acquiring an image of a crop during harvesting by an agricultural vehicle with an imaging device coupled to the agricultural vehicle, wherein the imaging device is a high-speed camera and acquiring the image of the crop occurs as the agricultural vehicle traverses a field including the crop;
analyzing the image of the crop identify portions of the image corresponding to leaves and identify portions of the image corresponding to stems;
generating a first metric indicating an amount of leaf in the image of the crop;
generating a second metric indicating an amount of stem in the image of the crop;
determining a leaf-to-stem ratio for the crop based on the first and second metric; and
signaling an agricultural machine to perform an action subsequent to harvesting based on the leaf-to-stem ratio, wherein the agricultural machine includes a baler.
11 . The crop analysis method of claim 10 , wherein analyzing the image of the crop includes applying a plurality of markers on the image.
12 . The crop analysis method of claim 11 , wherein the first metric is a count of markers overlaying leaves and the second metric is a count of markers overlaying stems.
13 . The crop analysis system of claim 10 , wherein the first metric is an image area measurement corresponding to leaves and the second metric is an image area measurement corresponding to stems.
14 . The crop analysis system of claim 10 , further comprising acquiring a plurality of images via the imaging device and determining an aggregate leaf-to-stem ratio corresponding to the plurality of images.