Grape Leaves Disease Detection

December 2, 2022

Built a Deep Learning model using TensorFlow and Keras for grape leaf disease detection

Grape Leaves Disease Detection — *Photo by* *Sasha Kolesnik* on *Dribbble*

Aim

To develop a Deep Learning model to identify diseased grape leaves using python programming language and Deep learning libraries

Prerequisites

Before getting started, you should have a good understanding of:

Python programming language
Deep Learning Libraries(Tensorflow, Keras, OpenCV)

Dataset

Link to download the dataset:

URL: https://cainvas-static.s3.amazonaws.com/media/user_data/vomchaithany/Grapes-Leaf-Disease-data.zip

get and unzip the data

	!wget -N "https://cainvas-static.s3.amazonaws.com/media/user_data/vomchaithany/Grapes-Leaf-Disease-data.zip"
	!unzip -o Grapes-Leaf-Disease-data.zip
	!rm Grapes-Leaf-Disease-data.zip

view raw gistfile1.txt hosted with ❤ by GitHub

output:

--2021-06-28 13:36:55--  https://cainvas-static.s3.amazonaws.com/media/user_data/vomchaithany/Grapes-Leaf-Disease-data.zip
Resolving cainvas-static.s3.amazonaws.com (cainvas-static.s3.amazonaws.com)... 52.219.64.72
Connecting to cainvas-static.s3.amazonaws.com (cainvas-static.s3.amazonaws.com)|52.219.64.72|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 23864365 (23M) [application/zip]
Saving to: ‘Grapes-Leaf-Disease-data.zip’

Grapes-Leaf-Disease 100%[===================>]  22.76M  81.8MB/s    in 0.3s

2021-06-28 13:36:55 (81.8 MB/s) - ‘Grapes-Leaf-Disease-data.zip’ saved [23864365/23864365]

Preparing data for the model

we assign train data to dirtrain and test data to dirtest

	dirtrain = 'train'
	dirtest = 'test'

view raw gistfile1.txt hosted with ❤ by GitHub

we are going to deal with 4 categories :

Black_rot
Esca_(Black_Measles)
Health
Leaf_blight_(Isariopsis_Leaf_Spot)

	for c in categories:
	path=os.path.join(dirtrain,c)
	for i in os.listdir(path):
	img_array=cv2.imread(os.path.join(path,i))
	print(img_array.shape)
	plt.imshow(img_array)
	plt.show()
	break
	break

view raw gistfile1.txt hosted with ❤ by GitHub

output:

Converting the images into arrays and storing them with labels

	#train data
	training_data = []
	def create_training_data():
	count = []
	for c in categories:
	path = os.path.join(dirtrain,c)
	class_num = categories.index(c)
	c = 0
	for i in os.listdir(path):
	c = c+1
	try:
	img_array = cv2.imread(os.path.join(path,i))
	training_data.append([img_array,class_num])
	except Exception as e:
	pass
	count.append(c)
	return count
	count_train=create_training_data()

	#test data
	testing_data = []
	def create_testing_data():
	count=[]
	for c in categories:
	path = os.path.join(dirtest,c)
	class_num = categories.index(c)
	c = 0
	for i in os.listdir(path):
	c = c+1
	try:
	img_array = cv2.imread(os.path.join(path,i))
	testing_data.append([img_array,class_num])
	except Exception as e:
	pass
	count.append(c)
	return count
	count_test=create_testing_data()

view raw gistfile1.txt hosted with ❤ by GitHub

shuffling the data

	random.shuffle(training_data)
	random.shuffle(testing_data)

view raw gistfile1.txt hosted with ❤ by GitHub

Resizing the images

	x_train = []
	y_train = []
	x_test = []
	y_test = []

view raw gistfile1.txt hosted with ❤ by GitHub

	for features, label in training_data:
	x_train.append(features)
	y_train.append(label)'
	x_train=np.array(x_train).reshape(-1,256,256,3)

	for features, label in testing_data:
	x_test.append(features)
	y_test.append(label)
	x_test = np.array(x_test).reshape(-1,256,256,3)

	x=cv2.resize(training_data[0][0],(256,256))
	plt.imshow(x,cmap='gray')

view raw gistfile1.txt hosted with ❤ by GitHub

output:

Creating the pickle files

	def save_training_data(x_train,y_train):
	pickle_out=open("x_train_coloured.pickle","wb")
	pickle.dump(x_train,pickle_out)
	pickle_out.close()

	pickle_out=open("y_train_coloured.pickle","wb")
	pickle.dump(y_train,pickle_out)
	pickle_out.close
	save_training_data(x_train,y_train)

	def save_testing_data(x_test,y_test):
	pickle_out=open("x_test_coloured.pickle","wb")
	pickle.dump(x_test,pickle_out)
	pickle_out.close()

	pickle_out=open("y_test_coloured.pickle","wb")
	pickle.dump(y_test,pickle_out)
	pickle_out.close()
	save_testing_data(x_test,y_test)

	def load_data():
	pickle_in = open("x_train_coloured.pickle","rb")
	x_train = pickle.load(pickle_in)
	return x_train

view raw gistfile1.txt hosted with ❤ by GitHub

Model

	K.clear_session()
	model = Sequential()
	model.add(layers.Conv2D(32,(3,3),padding='same',input_shape=(256,256,3),activation='relu'))
	model.add(layers.Conv2D(32,(3,3),activation='relu'))


	model.add(layers.MaxPool2D(pool_size=(8,8)))

	model.add(layers.Conv2D(32,(3,3),padding='same',activation='relu'))
	model.add(layers.Conv2D(32,(3,3),activation='relu'))

	model.add(layers.MaxPool2D(pool_size=(8,8)))

	model.add(Activation('relu'))

	model.add(Flatten())
	model.add(layers.Dense(256,activation='relu'))
	model.add(layers.Dense(4,activation='softmax'))

	model.compile(loss = 'categorical_crossentropy',optimizer = 'rmsprop',metrics = ['accuracy'])

	model.summary()

view raw gistfile1.txt hosted with ❤ by GitHub

output:

Model: “sequential”
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv2d (Conv2D) (None, 256, 256, 32) 896
_________________________________________________________________
conv2d_1 (Conv2D) (None, 254, 254, 32) 9248
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 31, 31, 32) 0
_________________________________________________________________
conv2d_2 (Conv2D) (None, 31, 31, 32) 9248
_________________________________________________________________
conv2d_3 (Conv2D) (None, 29, 29, 32) 9248
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 3, 3, 32) 0
_________________________________________________________________
activation (Activation) (None, 3, 3, 32) 0
_________________________________________________________________
flatten (Flatten) (None, 288) 0
_________________________________________________________________
dense (Dense) (None, 256) 73984
_________________________________________________________________
dense_1 (Dense) (None, 4) 1028
=================================================================
Total params: 103,652
Trainable params: 103,652
Non-trainable params: 0
_________________________________________________________________

	y_train_cat = to_categorical(y_train,4)
	y_test_cat = to_categorical(y_test,4)

	history = model.fit(x_train, y_train_cat, batch_size=32, epochs=10, verbose=1, validation_split=0.15, shuffle=True)

view raw gistfile1.txt hosted with ❤ by GitHub

output:

Epoch 1/10
 2/42 [>.............................] - ETA: 4s - loss: 33.4274 - accuracy: 0.4062WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0359s vs `on_train_batch_end` time: 0.0988s). Check your callbacks.
42/42 [==============================] - 6s 140ms/step - loss: 4.3365 - accuracy: 0.4440 - val_loss: 5.3007 - val_accuracy: 0.3276
Epoch 2/10
42/42 [==============================] - 6s 135ms/step - loss: 1.2223 - accuracy: 0.5103 - val_loss: 2.2340 - val_accuracy: 0.4095
Epoch 3/10
42/42 [==============================] - 6s 134ms/step - loss: 0.8702 - accuracy: 0.6398 - val_loss: 1.6332 - val_accuracy: 0.4698
Epoch 4/10
42/42 [==============================] - 5s 120ms/step - loss: 0.6871 - accuracy: 0.7380 - val_loss: 0.5808 - val_accuracy: 0.7198
Epoch 5/10
42/42 [==============================] - 6s 133ms/step - loss: 0.5256 - accuracy: 0.7761 - val_loss: 0.7016 - val_accuracy: 0.7457
Epoch 6/10
42/42 [==============================] - 6s 133ms/step - loss: 0.4148 - accuracy: 0.8378 - val_loss: 0.8888 - val_accuracy: 0.7155
Epoch 7/10
42/42 [==============================] - 6s 135ms/step - loss: 0.3828 - accuracy: 0.8462 - val_loss: 0.6643 - val_accuracy: 0.7931
Epoch 8/10
42/42 [==============================] - 5s 123ms/step - loss: 0.2999 - accuracy: 0.8919 - val_loss: 0.3769 - val_accuracy: 0.8664
Epoch 9/10
42/42 [==============================] - 5s 125ms/step - loss: 0.2474 - accuracy: 0.9177 - val_loss: 0.3600 - val_accuracy: 0.9009
Epoch 10/10
42/42 [==============================] - 5s 130ms/step - loss: 0.1429 - accuracy: 0.9528 - val_loss: 0.2745 - val_accuracy: 0.8966

Save the model

	# save the model
	model.save("leaf_disease_coloured.h5")
	new_model = tf.keras.models.load_model("leaf_disease_coloured.h5")

view raw gistfile1.txt hosted with ❤ by GitHub

Loss and Accuracy of the model

	loss, acc = new_model.evaluate(x_test,y_test_cat, verbose=2)
	print('Restored model, accuracy: {:5.2f}%'.format(100*acc))

view raw gistfile1.txt hosted with ❤ by GitHub

output:

18/18 - 1s - loss: 0.1902 - accuracy: 0.9304
Restored model, accuracy: 93.04%

The Metrics

	# train_loss vs Val_loss
	from matplotlib import pyplot as plt
	plt.plot(history.history['loss'])
	plt.plot(history.history['val_loss'])
	plt.title('model loss')
	plt.ylabel('loss')
	plt.xlabel('epoch')
	plt.legend(['train', 'validation'], loc='upper left')
	plt.show()

view raw gistfile1.txt hosted with ❤ by GitHub

output:

	# train_accuracy vs val_accuracy
	from matplotlib import pyplot as plt
	plt.plot(history.history['accuracy'])
	plt.plot(history.history['val_accuracy'])
	plt.title('model accuracy')
	plt.ylabel('acc')
	plt.xlabel('epoch')
	plt.legend(['train', 'validation'], loc='upper left')
	plt.show()

view raw gistfile1.txt hosted with ❤ by GitHub

output:

Custom Predictions

	img = cv2.resize(x_test[0],(256,256))
	plt.imshow(img)

view raw gistfile1.txt hosted with ❤ by GitHub

output:

	img = img.reshape(-1,256,256,3)
	predict_class = new_model.predict_classes(img)
	categories[predict_class[0]]

view raw gistfile1.txt hosted with ❤ by GitHub

output:

\'Healthy\'

Cainvas Gallery: https://cainvas.ai-tech.systems/gallery/

Conclusion

We’ve trained our Deep Learning model using TensorFlow and Keras for detecting grape leaf disease and got an accuracy of 93%.

Notebook link is here

Credit: Om Chaithanya V

Aim

Prerequisites

Dataset

Preparing data for the model

Converting the images into arrays and storing them with labels

Creating the pickle files

Model

Loss and Accuracy of the model

The Metrics

Custom Predictions

Conclusion

105 Serra Way, 251, Milpitas, CA.