Devin's repositories
DeformableCNN-PlantTraits
A deformable CNN model that accepts multiple sensor inputs and predicts multiple continuous plant trait outputs. SOTA on the 2021 Autonomous Greenhouse Challenge dataset.
leaf-area-counter
This code is designed to count and quantify leaves in an image. The best way to do this is to lay the leaves flat on a piece of paper with a blue disk of known size in the upper left hand quarter. Do not allow any of the leaves to extend above or to the left of the disk so that it is always counted as 1. This code uses red, green, and blue channels to identify and quantify the area of leaves. These channels can be separated or stacked.
unsupervised_x-rayct_image_cropping_tool
cropping almond bud images
3d_data_extractor_and_stack_viewer
Stack your 2d x-ray ct data, create new 3d tiff file, extract 3d data, and view 3d data with this code.
image_stacker_and_reslicer
Code to stack a series of 2d images into a 3d object based on image number and then reslice the 3d stack into a sequence of 2d images from any direction.
jetson-inference
Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.
root_area_counter
This code is designed to count and quantify roots in an image. The best way to do this is to lay the roots flat on a piece of paper with a blue disk of known size in the upper left hand quarter. Do not allow any of the roots to extend above or to the left of the disk so that it is always counted as 1. This code uses red, green, and blue channels to identify and quantify the area of roots. These channels can be separated or stacked.
vision
Datasets, Transforms and Models specific to Computer Vision
Visualizing_3d_structures
Visualizing labels in 3d