routing.py

from flask import render_template, request, redirect, url_for, session
import os
import torch
import fastai
from Imageclassifier import ImageClassifier 
import matplotlib.cm as cm
import numpy as np
from PIL import Image
import PIL


class Routing:
	def __init__(self, app, model, model1):
		self.app = app
		self.model = model
		self.model1 = model1

	def AppRouting(self):
		@self.app.route('/')
		def home():
			return render_template('index1.html')

		@self.app.route("/classify", methods = ["POST", "GET"])
		def classify():
			return render_template('appclassify.html')

		@self.app.route('/classify_image', methods = ["POST", "GET"])
		def classify_image():
			image2 = request.files.get('image2')
			image2.save(os.path.join(self.app.config['UPLOAD_FOLDER'], image2.filename))
			file_path2 = os.path.join(self.app.config['UPLOAD_FOLDER'], image2.filename)
			classifier = ImageClassifier()
			preds = classifier.model_predict(file_path2, self.model)
			index_tensor = torch.from_numpy(preds.numpy()).unsqueeze(-1)
			class_names2 = [classifier.classes[index] for index in index_tensor]
			session["class_names2"] = class_names2
			return redirect(url_for('classify_display', filename2 = image2.filename))

		@self.app.route('/classify_display/<filename2>')
		def classify_display(filename2):
			class_name2 = session.get("class_names2", None)
			class_name2 = class_name2[0]
			return render_template('appclassifydisplay.html', 
						image2 = url_for('static', filename='uploads/' + filename2),
						class_name2 = class_name2)



		@self.app.route("/segment", methods = ['POST', 'GET'])
		def segment():
			return render_template('appsegment.html')


		@self.app.route('/segment_display', methods=['GET', 'POST'])
		def upload_file():
			if request.method == 'POST':
				image1 = request.files.get('image1')
				image2 = request.files.get('image2')

				date1 = str(request.form['PastDate'])
				date2 = str(request.form['RecentDate'])

				date1 = date1[:-6] + " " + date1[-5:]
				date2 = date2[:-6] + " " + date2[-5:]
				# Save the files to the uploads folder in the static directory
				image1.save(os.path.join(self.app.config['UPLOAD_FOLDER'], image1.filename))
				image2.save(os.path.join(self.app.config['UPLOAD_FOLDER'], image2.filename))

				file_path2 = os.path.join(self.app.config['UPLOAD_FOLDER'], image2.filename)
				file_path1 = os.path.join(self.app.config['UPLOAD_FOLDER'], image1.filename)

				# segmentation part
				predict = self.model1.predict(file_path1)
				print(file_path1)
				pred_mask = predict[0]
				cmap = cm.get_cmap('viridis', np.max(pred_mask.numpy()) + 1)
				pred_mask_img = PIL.Image.fromarray((cmap(pred_mask.numpy())*255).astype(np.uint8)).convert("RGB")
				filename3 = 'image3.png'
				save_path1 = os.path.join(self.app.config['UPLOAD_FOLDER1'], filename3)
				pred_mask_img.save(save_path1)

				forest = np.sum(pred_mask.numpy() == 6)
				# Calculate the total area of the image
				total_area = predict[0].shape[0] * predict[0].shape[1]
				# Calculate the segmented area1
				forested_area1 = forest/ total_area * 100
				deforested_area1 = 100 - forested_area1

				predict1 = self.model1.predict(file_path2)
				print(file_path2)
				pred_mask1 = predict1[0]
				cmap = cm.get_cmap('viridis', np.max(pred_mask1.numpy()) + 1)
				pred_mask_img1 = PIL.Image.fromarray((cmap(pred_mask1.numpy())*255).astype(np.uint8)).convert("RGB")  
				filename4 = 'image4.png'
				save_path2 = os.path.join(self.app.config['UPLOAD_FOLDER1'], filename4)
				pred_mask_img1.save(save_path2)

				forest = np.sum(pred_mask1.numpy()== 6)
				total_area = predict[0].shape[0] * predict[0].shape[1]
				forested_area2 = (forest/ total_area) * 100
				deforested_area2 = 100 - forested_area2
				value = deforested_area2 - deforested_area1
				if value >0:
				    string = 'deforested'
				else:
				    string = 'forested'
				value = round(value, 2)
				value = abs(value)
				session["filename1"] = image1.filename
				session["filename2"] = image2.filename
				session["filename3"] = filename3
				session["filename4"] = filename4
				session["date1"] = date1
				session["date2"] = date2
				session["value"] = value
				session["string"] = string

				return redirect(url_for('segment_display', filename1 = image1.filename, filename2 = image2.filename, filename3 = filename3, filename4 = filename4))
			return '''
		    
		    '''



		@self.app.route('/segment_images/<filename1>/<filename2>/<filename3>/<filename4>')
		def segment_display(filename1, filename2, filename3, filename4):
		    value = session.get("value", None)
		    date1 = session.get("date1", None)
		    date2 = session.get("date2", None)
		    string = session.get("string", None)
		    return render_template('appsegmentdisplay.html', image1 = url_for('static', filename='uploads/' + filename1), 
		                                                   image2 = url_for('static', filename='uploads/' + filename2),
		                                                   image3 = url_for('static', filename='Segments/' + filename3),
		                                                   image4 = url_for('static', filename='Segments/' + filename4),
		                                                   date1 = date1, date2 = date2,
		                                                    value = value,
		                                                    string = string
		                                                   )


		@self.app.route("/previous", methods = ["POST"])
		def previous():
			return redirect(url_for('segment'))


		@self.app.route("/previous1", methods = ["POST"])
		def previous1():
		    return redirect(url_for('classify'))


		@self.app.route("/redirect", methods=["POST"])
		def next_picture():
		    return redirect(url_for("home"))