Merge pull request #76 from PUTvision/devel

devel->master

.github/workflows/python-app.yml → .github/workflows/python-app-ubuntu.yml

@@ -11,7 +11,11 @@ on:
 
 jobs:
   build:
-    runs-on: ubuntu-22.04
+    strategy:
+      matrix:
+        os: [ubuntu-20.04, ubuntu-22.04]
+
+    runs-on: ${{ matrix.os }}
 
     steps:
     - uses: actions/checkout@v3
@@ -22,6 +26,13 @@ jobs:
     - name: Update
       run: |
         sudo apt update
+    - if: matrix.os == 'ubuntu-20.04'
+      run: |
+        sudo apt install gnupg software-properties-common
+        sudo mkdir -m755 -p /etc/apt/keyrings
+        sudo wget -O /etc/apt/keyrings/qgis-archive-keyring.gpg https://download.qgis.org/downloads/qgis-archive-keyring.gpg
+        echo $'Types: deb deb-src\nURIs: https://qgis.org/debian\nSuites: focal\nArchitectures: amd64\nComponents: main\nSigned-By: /etc/apt/keyrings/qgis-archive-keyring.gpg' | sudo tee -a /etc/apt/sources.list.d/qgis.sources
+        sudo apt update
     - name: Install qgis
       run: |
         sudo apt install -yq qgis

README.md

@@ -4,7 +4,7 @@
   <h2 align="center">Deepness: Deep Neural Remote Sensing QGIS Plugin</h2>
 </p>
 
-![main](https://github.com/PUTvision/qgis-plugin-deepness/actions/workflows/python-app.yml/badge.svg)
+![main](https://github.com/PUTvision/qgis-plugin-deepness/actions/workflows/python-app-ubuntu.yml/badge.svg)
 [![GitHub contributors](https://img.shields.io/github/contributors/PUTvision/qgis-plugin-deepness)](https://github.com/PUTvision/qgis-plugin-deepness/graphs/contributors)
 [![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg?style=flat-square)](https://makeapullrequest.com)
 [![GitHub stars](https://img.shields.io/github/stars/PUTvision/qgis-plugin-deepness)](https://github.com/PUTvision/qgis-plugin-deepness/stargazers)

docs/source/example/example_segmentation_landcover.rst

@@ -22,7 +22,8 @@ We built our pipeline based on `Pytorch <https://github.com/pytorch/pytorch>`_,
 * Model :code:`DeepLabV3+` with :code:`tu-semnasnet_100` backend
 * Loss function: balanced :code:`FocalDiceLoss`
 * Input image size: :code:`512x512`
-* Normalization: :code:`mean=[0.485, 0.456, 0.406]`, :code:`std=[0.229, 0.224, 0.225]`
+* Normalization: :code:`1./255.`
+* Color order: :code:`RGB`
 
 ==================
 Converting to onnx

docs/source/main/main_supported_versions.rst

@@ -14,11 +14,11 @@ Supported Operating Systems
 +---------------+------------------------------------+
 | OS Family     |  Operating System                  |
 +===============+====================================+
-| Linux         |  Ubuntu 22.04, Fedora 37*          |
+| Linux         |  Ubuntu 20.04/22.04, Fedora 37*    |
 +---------------+------------------------------------+
 | Windows       |  Windows 10                        |
 +---------------+------------------------------------+
-| Mac OS        |                                    |
+| Mac OS        |  Big Sur                           |
 +---------------+------------------------------------+
 
 :code:`*` means that it should work but has not beed tested yet
@@ -28,4 +28,5 @@ The plug-in was tested in the environment:
 
   - Ubuntu 22.04.1 LTS and QGIS 3.22.11
   - Ubuntu 22.04.1 LTS and QGIS 3.22.4
+  - Ubuntu 20.04 LTS and QGIS 3.28
   - Windows 10.0.19043 and QGIS 3.26.3

docs/source/main/model_zoo/MODEL_ZOO.md

@@ -20,16 +20,20 @@ The [Model ZOO](https://chmura.put.poznan.pl/s/2pJk4izRurzQwu3) is a collection
 ## Object detection models
 
 | Model name | Input size | CM/PX | Description | Example image |
-|---|---|---|---|
+|---|---|---|---|---|
 | [Airbus Planes Detection](https://chmura.put.poznan.pl/s/bBIJ5FDPgyQvJ49) | 256 | 70 | YOLOv7 tiny model for object detection on satellite images. Based on the [Airbus Aircraft Detection dataset](https://www.kaggle.com/datasets/airbusgeo/airbus-aircrafts-sample-dataset). | [Image](https://chmura.put.poznan.pl/s/VfLmcWhvWf0UJfI) |
 | [Airbus Oil Storage Detection](https://chmura.put.poznan.pl/s/gMundpKsYUC7sNb) | 512 | 150 | YOLOv5-m model for object detection on satellite images. Based on the [Airbus Oil Storage Detection dataset](https://www.kaggle.com/datasets/airbusgeo/airbus-oil-storage-detection-dataset). | [Image](https://chmura.put.poznan.pl/s/T3pwaKlbFDBB2C3) |
 
 ## Contributing
 
-* PRs with models are welcome! Please follow the [general model information](https://qgis-plugin-deepness.readthedocs.io/en/latest/creators/creators_description_classes.html).
+PRs with models are welcome! 
+
+* Please follow the [general model information](https://qgis-plugin-deepness.readthedocs.io/en/latest/creators/creators_description_classes.html).
 
 * Use `MODEL_ZOO` tag in your PRs to make it easier to find them.
 
 * If you need, you can check [how to export the model to ONNX](https://qgis-plugin-deepness.readthedocs.io/en/latest/creators/creators_example_onnx_model.html).
 
-* And do not forget to [add metadata to the ONNX model](https://qgis-plugin-deepness.readthedocs.io/en/latest/creators/creators_add_metadata_to_model.html). You can host your model yourself or ask us to do it.
+* And do not forget to [add metadata to the ONNX model](https://qgis-plugin-deepness.readthedocs.io/en/latest/creators/creators_add_metadata_to_model.html). 
+
+* You can host your model yourself or ask us to do it.

plugin/deepness/dialogs/packages_installer/packages_installer_dialog.py

@@ -6,6 +6,7 @@
 import importlib
 import logging
 import os
+import platform
 import subprocess
 import sys
 import traceback
@@ -48,8 +49,15 @@ def __str__(self):
 # NOTE! For the time being requirement are repeated as in `requirements.txt` file
 # Consider merging it into some common file in the future
 # (if we can use a fixed version of pip packages for all python versions)
+
+opencv_version = '4.6.0.66'
+if sys.platform == "linux" or sys.platform == "linux2":
+    import lsb_release
+    if lsb_release.get_os_release()['CODENAME'] == 'focal':
+        opencv_version = '4.5.5.64'
+
 packages_to_install = [
-    PackageToInstall(name='opencv-python-headless', version='4.6.0.66', import_name='cv2'),
+    PackageToInstall(name='opencv-python-headless', version=opencv_version, import_name='cv2'),
 ]
 
 onnx_runtime_version = '1.12.1'

plugin/deepness/metadata.txt

@@ -4,9 +4,9 @@
 
 [general]
 name=Deepness: Deep Neural Remote Sensing
-qgisMinimumVersion=3.10.14
+qgisMinimumVersion=3.22
 description=Inference of deep neural network models (ONNX) for segmentation, detection and regression
-version=0.4.0
+version=0.4.1
 author=PUT Vision
 [email protected]
 

plugin/deepness/processing/models/detector.py

@@ -207,8 +207,6 @@ def xywh2xyxy(x: np.ndarray) -> np.ndarray:
         y[:, 3] = x[:, 1] + x[:, 3] / 2  # bottom right y
         return y
 
-    @staticmethod
-
     @staticmethod
     def non_max_suppression_fast(boxes: np.ndarray, probs: np.ndarray, iou_threshold: float) -> List:
         """Apply non-maximum suppression to bounding boxes

requirements.txt

@@ -3,6 +3,7 @@
 #pyqt5  # available already from qgis repo
 
 
-# NOTE - for the time being - keep the same packages and versions in the `package_installer_dialog.py`
+# NOTE - for the time being - keep the same packages and versions in the `packages_installer_dialog.py`
 onnxruntime-gpu==1.12.1
-opencv-python-headless==4.6.0.66  # opencv-python if you would like to display images while debugging
+opencv-python-headless==4.6.0.66; python_version > '3.9'  # opencv-python if you would like to display images while debugging
+opencv-python-headless==4.5.5.64; python_version <= '3.9'