Web app using Django and an Android app

.gitignore

@@ -1,8 +1,6 @@
 *.pyc
 /train/v1_test.tar.bz2
 /train/v1_training.tar.bz2
-.gitignore
-.gitignore.swp
 mnist.pkl.gz
 save
 scripts/validator.py

README.md

@@ -1,98 +1,57 @@
-SnapSudoku
-===================
+# SnapSudoku
 
-*Take a picture of a Sudoku and have SnapSudoku solve it for you!*
+Take a picture of an unsolved Sudoku and have SnapSudoku solve it for you! This repository also consists of an Android application which harnesses an API endpoint created using [Django REST Framework](http://www.django-rest-framework.org/).
 
+## How to use
 
-----------
- TODO:
----------
- - [ ] Improve algorithm to get better Sudoku Grid extraction, make it `more robust against blurs` .
- - [ ] Improve `empty cell detection`. Only a basic logic is used right now. 
+```
+$ git clone https://github.com/ymittal/SnapSudoku.git
+$ cd SnapSudoku
+$ pip install -r requirements.txt
+$ python snapsudoku/sudoku.py <path>
+```
 
-Prerequisites:
--------------
+You can read the directions to test the Android application [**here**](https://github.com/ymittal/SnapSudoku/blob/master/android/README.md).
 
-- Python 2.7 but __not__ Python 3
-    - Download from [here](https://www.python.org/downloads/)
+### Algorithm
 
-- OpenCV
-    - `sudo apt-get install python-opencv` (preferred)
-    - Install OpenCV from [here](http://opencv.org/downloads.html) 
+ - Basic image preprocessing - **Thresholding**
+ - Crop out approx. Sudoku puzzle (largest contour)
+ - Get the grid square verticles (*need for improvement*)
+    - Get the largest contour of the image
+    - Get the largest bounding rectangle within the contour
+    - Compute the grid corners
+ - Do a **warp perspective** on the Sudoku
+ - Extract cells by slicing the grid evenly
+ - Isolate cells as follows
+    - Extract the largest connected component, giving more priority to the center pixels
+    - Remove all major noise in a cell
+ - Predict digits using a Neural Network
 
-- Numpy (1.11.0)
-    - `pip install numpy` (preferred)
-    - You can build it from source [here](https://github.com/numpy/numpy)
+The Neural Network created was trained with around 250 digits. The [constants](https://github.com/ymittal/SnapSudoku/blob/master/snapsudoku/networks/net) used in the training phase and the [training dataset](https://github.com/ymittal/SnapSudoku/tree/master/snapsudoku/train) can be found in the repository itself.
 
-How to use: 
-----------
-    git clone https://github.com/prajwalkr/SnapSudoku.git
-    cd SnapSudoku
-    python sudoku.py <path-to-input-image>
+### Working
 
+Here's a Sudoku image from a smartphone:
 
-Working:
--------
-> Here's a Sudoku image from a smartphone:
+<img src="https://lh3.googleusercontent.com/-rGpsVTsqkYU/VwysnNV6U4I/AAAAAAAAG00/1XVCxhPkVzMTugwy53PUTVu76JtywthyQCLcB/s1000/test1.jpg" alt="Input Image" width="320px">
 
-![Input Sudoku Image](https://lh3.googleusercontent.com/-rGpsVTsqkYU/VwysnNV6U4I/AAAAAAAAG00/1XVCxhPkVzMTugwy53PUTVu76JtywthyQCLcB/s1000/test1.jpg "Input image")![](blob:https%3A//drive.google.com/7556d1d6-752d-4e86-b913-8373d50ebe41) 
-</br>
+The current code produces the following output:
 
-> The current code gives out the following output to the Terminal:
+<img src="https://lh3.googleusercontent.com/-PLL7mtKdT68/VxzRU-D1mxI/AAAAAAAAH4s/v2lvYd_mQes3J3ta0PwQ_W2gwxhrVS_VQCLcB/s500/Final+Result.png" alt="Final Result" width="320px">
 
-![Final result](https://lh3.googleusercontent.com/-PLL7mtKdT68/VxzRU-D1mxI/AAAAAAAAH4s/v2lvYd_mQes3J3ta0PwQ_W2gwxhrVS_VQCLcB/s500/Final+Result.png "Final Result")
+**Note:** For a detailed step-by-step demo, check out [**this**](https://github.com/ymittal/SnapSudoku/blob/master/snapsudoku/demo/README.md).
 
-Algorithm
--------------
+## TODO
 
- > 1. Basic image preprocessing - **Thresholding**.
- > 2. Crop out approx. Sudoku puzzle (Largest contour)
- > 3. Get the grid square vertices: *(a better way of doing this is required!)* </br>
-   3.1. Get the **largest contour** of the image.</br>
-   3.2. Get the largest **bounding rectangle** within the contour.</br>
-   3.3. Compute the grid corners. 
-> 4. Do a **Warp perspective** on the sudoku image
-> 5. We will extract cells from this, by slicing the sudoku grid evenly.
-> 6. Digit isolation in cell is done through a series of steps: </br>
-    6.1. Extract the **largest connected component** in the image, giving more *priority to the center pixels*. </br>
-    6.2. Removing all major noise in the cell. 
-> 7.  Predict Digits using a Neural Network. 
-
-> The only 3rd party libraries required are  *OpenCV, Numpy*. The Neural Network created was trained with around 250 digits. The constants used in the training phase, the training data-set is in this repository itself. 
-
-Here are some illustrations of the different stages:
--------
-
-> After Preprocessing:
-
-![After Preprocessing](https://lh3.googleusercontent.com/-hTPN4mSDNiY/Vwy8UgTcxNI/AAAAAAAAG1c/e67gE9TSAKQrcd-ADHmAgOtuMDQPhyCrgCLcB/s500/After+Preprocessing.png "After Preprocessing")
-
-> Final processed Sudoku Grid
-
-![enter image description here](https://lh3.googleusercontent.com/-AcbLo77wYH0/VxtPrbXcAYI/AAAAAAAAH0E/OmEzl2Hn9JkjQTxfdAVTn7ZeN3q3rsutgCLcB/s500/Sudoku+Grid.png "Sudoku Grid Image")
-
-> Here are a few digits after processing the cells:
-
-![Digit 3](https://lh3.googleusercontent.com/-2zxwex6LYnk/VxtQEQH6fDI/AAAAAAAAH0Q/nFmrwIlm7HYV7O2qZEoICQDKF8fcoFKmQCLcB/s100/three.png "Three")![Digit 8](https://lh3.googleusercontent.com/-oPRnuu7XXxc/VwzHKJjnpyI/AAAAAAAAG2o/11FlxwHkkygGEHgoY4NQLZroq-fH6b5MACLcB/s100/eight.png "eight.png") 
-
-> Here's a typical empty cell:
-
-![empty cell](https://lh3.googleusercontent.com/-p2bhyuRWptI/VwzHVYrtABI/AAAAAAAAG2w/C_vKYzb75sQ8gcPdf0aaHCjB6dM02du8wCLcB/s100/emptycell.png "emptycell.png")
-
->  Predicted Grid:
-
-![Prediction](https://lh3.googleusercontent.com/-bmiUuMHZtYw/VxtQ8SezWLI/AAAAAAAAH0s/VV8HvATwHEAfhtKJqT6nK-fh0A28E52gwCLcB/s500/Digits.png "Digits")
-
-> Solved Grid:
-
-![Solved Grid](https://lh3.googleusercontent.com/-zWnlGKCz9xs/VxzSmieN5EI/AAAAAAAAH48/DR6t33TSfhgDVe7ew3n0YRKAUJB45rKWACLcB/s500/Solved+Grid.png "Solved Grid")
-----------
-Contributors
-------------------
-[cclauss](https://github.com/cclauss) <br/>
-[lakshmanaram](https://github.com/lakshmanaram)
-
-I'll be very happy to get new ideas to improve the accuracy and make the application better. Feel free to give a pull request! :smile:
+- [ ] Improve algorithm to get better Sudoku grid extraction and make it more robust against blurs.
+- [ ] Improve empty cell detection. Only a basic logic is being used right now. 
 
+### Contributors
 
+* [prajwalkr](https://github.com/prajwalkr) **Prajwal Renukanand**
+* [cclauss](https://github.com/cclauss)
+* [lakshmanaram](https://github.com/lakshmanaram)
+* [ymittal](https://github.com/ymittal) **Yash Mittal**
 
+We are happy to get new ideas to make the application more accurate. Feel free to submit a pull request or open an issue. :smile:

algorithm.md

@@ -0,0 +1,14 @@
+### Algorithm
+
+ - Basic image preprocessing - **Thresholding**
+ - Crop out approx. Sudoku puzzle (largest contour)
+ - Get the grid square verticles (*need for improvement*)
+    - Get the largest contour of the image
+    - Get the largest bounding rectangle within the contour
+    - Compute the grid corners
+ - Do a **warp perspective** on the Sudoku
+ - Extract cells by slicing the grid evenly
+ - Isolate cells as follows
+    - Extract the largest connected component, giving more priority to the center pixels
+    - Remove all major noise in a cell
+ - Predict digits using a Neural Network

android/.gitignore

@@ -0,0 +1,9 @@
+*.iml
+.gradle
+/local.properties
+/.idea/workspace.xml
+/.idea/libraries
+.DS_Store
+/build
+/captures
+.externalNativeBuild

android/README.md

@@ -0,0 +1,18 @@
+# SnapSudoku Android
+
+An Android application to solve an unsolved sudoku by taking its picture
+
+## Getting Started
+
+* Before testing on an emulator, run the Django server as follows. You can access `localhost` at `http://10.0.2.2:8000/` on an Android emulator.
+
+  ```
+  $ python manage.py runserver
+  ```
+  **Note:** You need to turn on camera settings for your emulator.
+
+* If you deployed the Django app, add the URL to the list of `ALLOWED_HOSTS` in [`settings.py`](https://github.com/ymittal/SnapSudoku/blob/master/django_project/settings.py). Additionally, modify the `baseUrl` of `Retrofit.Builder()` in the Java code.
+
+## Authors
+
+Please feel free to reach out to [Yash Mittal]([email protected]) if you have any questions regarding the Android codebase.

android/app/.gitignore

@@ -0,0 +1 @@
+/build

android/app/build.gradle

@@ -0,0 +1,32 @@
+apply plugin: 'com.android.application'
+
+android {
+    compileSdkVersion 25
+    buildToolsVersion "25.0.1"
+    defaultConfig {
+        applicationId "com.example.yash.snapsudoku"
+        minSdkVersion 19
+        targetSdkVersion 25
+        versionCode 1
+        versionName "1.0"
+        testInstrumentationRunner "android.support.test.runner.AndroidJUnitRunner"
+    }
+    buildTypes {
+        release {
+            minifyEnabled false
+            proguardFiles getDefaultProguardFile('proguard-android.txt'), 'proguard-rules.pro'
+        }
+    }
+}
+
+dependencies {
+    compile fileTree(dir: 'libs', include: ['*.jar'])
+    androidTestCompile('com.android.support.test.espresso:espresso-core:2.2.2', {
+        exclude group: 'com.android.support', module: 'support-annotations'
+    })
+    compile 'com.android.support:appcompat-v7:25.0.1'
+    testCompile 'junit:junit:4.12'
+
+    compile 'com.squareup.retrofit2:retrofit:2.1.0'
+    compile 'com.squareup.retrofit2:converter-gson:2.0.0-beta4'
+}

android/app/proguard-rules.pro

@@ -0,0 +1,17 @@
+# Add project specific ProGuard rules here.
+# By default, the flags in this file are appended to flags specified
+# in C:\Users\Yash\AppData\Local\Android\Sdk/tools/proguard/proguard-android.txt
+# You can edit the include path and order by changing the proguardFiles
+# directive in build.gradle.
+#
+# For more details, see
+#   http://developer.android.com/guide/developing/tools/proguard.html
+
+# Add any project specific keep options here:
+
+# If your project uses WebView with JS, uncomment the following
+# and specify the fully qualified class name to the JavaScript interface
+# class:
+#-keepclassmembers class fqcn.of.javascript.interface.for.webview {
+#   public *;
+#}