Merge pull request #247 from sbrunner/dev-images

Add image to understand the deskewing

.github/workflows/main.yaml

@@ -41,7 +41,7 @@ jobs:
 
       - run: echo "${HOME}/.local/bin" >> ${GITHUB_PATH}
       - run: python3 -m pip install --user $(grep poetry ci/requirements.txt)
-      - run: poetry install
+      - run: poetry install --extras=debug_images
       - name: Versions
         run: |
           poetry --version
@@ -58,7 +58,8 @@ jobs:
       - uses: actions/upload-artifact@v3
         with:
           name: Test results
-          path: results/test-reports
+          path: results
+        if: always()
 
   build:
     name: Continuous integration
@@ -86,8 +87,8 @@ jobs:
       - name: Checks
         run: c2cciutils-checks
 
-      - run: poetry install
-      - run: poetry run prospector --output=pylint -X
+      - run: poetry install --extras=debug_images
+      - run: poetry run prospector --output=pylint --die-on-tool-error
 
       - name: Init Git
         run: |

.prospector.yaml

@@ -5,6 +5,7 @@ doc-warnings: True
 pylint:
   disable:
     - too-many-arguments
+    - too-many-statements
 
 pycodestyle:
   disable:
@@ -18,6 +19,7 @@ pydocstyle:
     - D102 # Missing docstring in public method
     - D104 # Missing docstring in public package
     - D212 # Multi-line docstring summary should start at the first line
+    - D202 # No blank lines allowed after function docstring
 
 bandit:
   run: true

README.md

@@ -79,4 +79,17 @@ rotated = rotate(image, angle, (0, 0, 0))
 cv2.imwrite('output.png', rotated)
 ```
 
+# Debug images
+
+If you get wrong skew angle you can generate debug images, that can help you to tune the skewing detection.
+
+If you install deskew with `pip install deskew[debug_images]` you can get some debug images used for
+the skew detection with the function `determine_skew_debug_images`.
+
+To start the investigation you should first increase the `num_peaks` (default `20`) and use
+the `determine_skew_debug_images` function.
+
+Then you can try to tune the following arguments `num_peaks`, `angle_pm_90`, `min_angle`, `max_angle`,
+`min_deviation` and eventually `sigma`.
+
 Inspired by Alyn: https://github.com/kakul/Alyn