All Projects → Olament → DeepMushroom

Olament / DeepMushroom

Licence: other
Image classification of Fungus using ResNet

Programming Languages

go
31211 projects - #10 most used programming language
python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to DeepMushroom

Iseebetter
iSeeBetter: Spatio-Temporal Video Super Resolution using Recurrent-Generative Back-Projection Networks | Python3 | PyTorch | GANs | CNNs | ResNets | RNNs | Published in Springer Journal of Computational Visual Media, September 2020, Tsinghua University Press
Stars: ✭ 202 (+215.63%)
Mutual labels:  resnet
Wideresnet Pytorch
Wide Residual Networks (WideResNets) in PyTorch
Stars: ✭ 249 (+289.06%)
Mutual labels:  resnet
Portrait FCN and 3D Reconstruction
This project is to convert PortraitFCN+ (by Xiaoyong Shen) from Matlab to Tensorflow, then refine the outputs from it (converted to a trimap) using KNN and ResNet, supervised by Richard Berwick.
Stars: ✭ 61 (-4.69%)
Mutual labels:  resnet
Deepfake Detection
Towards deepfake detection that actually works
Stars: ✭ 213 (+232.81%)
Mutual labels:  resnet
Pyramidnet Pytorch
A PyTorch implementation for PyramidNets (Deep Pyramidal Residual Networks, https://arxiv.org/abs/1610.02915)
Stars: ✭ 234 (+265.63%)
Mutual labels:  resnet
CNN-models
YOLO-v2, ResNet-32, GoogLeNet-lite
Stars: ✭ 32 (-50%)
Mutual labels:  resnet
Deep Learning With Python
Deep learning codes and projects using Python
Stars: ✭ 195 (+204.69%)
Mutual labels:  resnet
chainer-grad-cam
Chainer implementation of Grad-CAM
Stars: ✭ 20 (-68.75%)
Mutual labels:  resnet
Resnest
ResNeSt: Split-Attention Networks
Stars: ✭ 2,938 (+4490.63%)
Mutual labels:  resnet
Distributed-ResNet-Tensorflow
A Distributed ResNet on multi-machines each with one GPU card.
Stars: ✭ 20 (-68.75%)
Mutual labels:  resnet
R Centernet
detector for rotated-object based on CenterNet/基于CenterNet的旋转目标检测
Stars: ✭ 226 (+253.13%)
Mutual labels:  resnet
Fusenet
Deep fusion project of deeply-fused nets, and the study on the connection to ensembling
Stars: ✭ 230 (+259.38%)
Mutual labels:  resnet
MQBench Quantize
QAT(quantize aware training) for classification with MQBench
Stars: ✭ 29 (-54.69%)
Mutual labels:  resnet
Tensorflow Computer Vision Tutorial
Tutorials of deep learning for computer vision.
Stars: ✭ 206 (+221.88%)
Mutual labels:  resnet
Xtreme-Vision
A High Level Python Library to empower students, developers to build applications and systems enabled with computer vision capabilities.
Stars: ✭ 77 (+20.31%)
Mutual labels:  resnet
Sca Cnn.cvpr17
Image Captions Generation with Spatial and Channel-wise Attention
Stars: ✭ 198 (+209.38%)
Mutual labels:  resnet
Ai papers
AI Papers
Stars: ✭ 253 (+295.31%)
Mutual labels:  resnet
car-detection-model-prediction
No description or website provided.
Stars: ✭ 18 (-71.87%)
Mutual labels:  resnet
sharpmask
TensorFlow implementation of DeepMask and SharpMask
Stars: ✭ 31 (-51.56%)
Mutual labels:  resnet
mxnet-retrain
Create mxnet finetuner (retrain) for mac/linux ,no need install docker and supports CPU, GPU(eGpu/cudnn).support the inception,resnet ,squeeznet,mobilenet...
Stars: ✭ 32 (-50%)
Mutual labels:  resnet

DeepMushroom

DeepMushroom is a fungal classficiation project using ResNet

Data Sources

iNaturalist.org

iNaturalist.org is a citizen science website that allows people to upload images of unknown organisms for identification by other ecology enthusiasts. We collected images of identified fungi uploaded between 2017 and 2019 via the iNaturalist export tool. The images downloaded were then categorized by their species.

We use both python and golang scripts to download the images from iNaturalist.org. See here: main.go

Distribution

The data distribution is heavily skewed towards the few most common species. We remove the fungal species with less than 10 images for two reasons:

  • If one species has less than 10 identification on iNaturalist.org, it indicates that it is not frequently occuring. Therefore, there is less value in the identification of such species.
  • There is not enough data to effectively train the identification model. A species with less than 10 images will hurt the overall accuracy of our model

MushroomExpert.com

Since the images from MushroomExpert were identified by mycologists, we can use their images as a reliable validator to test the performance of our model.

Model

Since we are in the very early stage of the experiment we built the model with the fast.ai library. The model will gradually switch to our own models utilizing pytorch as we progress.

Metrics

Architecture Validation Accuracy Validation Top-5 Accuracy Test Accurarcy Test Top-5 Accuracy
ResNet34 70.68 86.36 31.94 48.11
ResNet50 79.67 91.76 38.77 59.14
ResNet50+Focal Loss 80.24 92.32 39.48 60.45

Top 10 Most Confused Fungal Species

Prediction Ground Truth
Fomitopsis mounceae Fomitopsis pinicola
Pleurotus pulmonarius Pleurotus ostreatus
Dacrymyces chrysospermus Tremella mesenterica
Tremella mesenterica Dacrymyces chrysospermus
Laetiporus gilbertsonii Laetiporus sulphureus
Stereum hirsutum Stereum complicatum
Tremella aurantia Tremella mesenterica
Ganoderma megaloma Ganoderma applanatum
Laetiporus cincinnatus Laetiporus sulphureus
Ganoderma applanatum Ganoderma brownii
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].