model added

sign-language/.vscode/settings.json

+{
+    "python.pythonPath": "/home/shreyansh/.conda/envs/ml725/bin/python"
+}
\ No newline at end of file

sign-language/main.py

@@ -7,15 +7,47 @@ cap = cv2.VideoCapture(0)
 model = tf.keras.models.load_model("models/model.h5")
 print("Loaded model from disk")
 
+sign = {
+0:"a",
+1:"b",
+2:"c",
+3:"d",
+4:"e",
+5:"f",
+6:"g",
+7:"h",
+8:"i",
+9:"k",
+10:"l",
+11:"m",
+12:"n",
+13:"o",
+14:"p",
+15:"q",
+16:"r",
+17:"s",
+18:"t",
+19:"u",
+20:"v",
+21:"w",
+22:"x",
+23:"y"
+}
+
 while(True):
     # Capture frame-by-frame
     ret, frame = cap.read()
-
-    # Our operations on the frame come here
+    cv2.rectangle(frame,(100,100),(300,300),(0,255,0),0)
     gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+    roi = gray[100:300, 100:300]
+    resized = cv2.resize(roi, (28,28), interpolation = cv2.INTER_AREA)
+    resized = resized.reshape(-1,28,28,1).astype(float)
+    print(sign[np.argmax(model.predict(resized))])
+    # Our operations on the frame come here
 
     # Display the resulting frame
     cv2.imshow('frame',gray)
+
     if cv2.waitKey(1) & 0xFF == ord('q'):
         break
 

sign-language/prediction.py

+import pandas as pd
+import numpy as np
+import tensorflow as tf
+from keras.models import Sequential,model_from_json,load_model
+from keras.layers import Dense, Conv2D, Flatten, MaxPooling2D, Dropout
+from sklearn.preprocessing import LabelBinarizer
+
+trainData = pd.read_csv("/content/drive/My Drive/Colab Notebooks/data/sign-language-mnist/sign_mnist_train.csv").values
+testData = pd.read_csv("/content/drive/My Drive/Colab Notebooks/data/sign-language-mnist/sign_mnist_test.csv").values
+
+trainX,trainY = trainData[:,1:],LabelBinarizer().fit_transform(trainData[:,0:1])
+testX,testY = testData[:,1:],LabelBinarizer().fit_transform(testData[:,0:1])
+
+trainX = np.divide(trainX,255)
+testX = np.divide(testX,255)
+
+# create model
+model = Sequential()
+model.add(Conv2D(64, kernel_size=(3,3), activation = 'relu', input_shape=(28, 28 ,1) ))
+model.add(MaxPooling2D(pool_size = (2, 2)))
+
+model.add(Conv2D(64, kernel_size = (3, 3), activation = 'relu'))
+model.add(MaxPooling2D(pool_size = (2, 2)))
+
+model.add(Conv2D(64, kernel_size = (3, 3), activation = 'relu'))
+model.add(MaxPooling2D(pool_size = (2, 2)))
+
+model.add(Flatten())
+model.add(Dense(128, activation = 'relu'))
+model.add(Dropout(0.35))
+model.add(Dense(24, activation = 'softmax'))
+
+model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
+
+trainX = trainX.reshape(trainX.shape[0], 28, 28, 1)
+testX = testX.reshape(testX.shape[0], 28, 28, 1)
+
+model.fit(trainX, trainY, validation_data=(testX, testY), epochs=40, batch_size=128)
+# Final evaluation of the model
+scores = model.evaluate(testX, testY, verbose=0)
+
+model.save("/content/drive/My Drive/Colab Notebooks/data/model.h5")
+
+print("CNN Error: %.2f%%" % (100-scores[1]*100))
\ No newline at end of file