Orb 1

Day-Night-Classifier/DayNight_Blue_Thresholds.py

@@ -21,12 +21,6 @@
 import numpy as np
 import matplotlib.pyplot as plt
 
-'''Training and Testing Data
-The 92 day/night images are separated into training and testing datasets.
-
-52 of these images are training images, for you to use as you create a classifier.
-40 are test images, which will be used to test the accuracy of your classifier.
-First, we set some variables to keep track of some where our images are stored'''
 
 '''First, we set some variables to keep track of some where our images are stored:'''
 image_dir_training = "day_night_images/training/"
@@ -70,20 +64,13 @@
 print("Label [1 = day, 0 = night]: " + str(selected_label))
 
 
-'''Feature Extraction
-Create a feature that represents the brightness in an image. 
-We'll be extracting the average brightness using HSV colorspace.
- Specifically, we'll use the V channel (a measure of brightness),
- add up the pixel values in the V channel, then divide that sum 
- by the area of the image to get the average Value of the image.
-
+'''
 Find the average brightness using the V channel
 This function takes in a standardized RGB image and returns
  a feature (a single value) that represent the average level of 
  brightness in the image. We'll use this value to classify
  the image as day or night.
 '''
-
 # Find the average Value or brightness of an image
 def avg_brightness(rgb_image):
     # Convert image to HSV

Day-Night-Classifier/DayNight_Classifier.py

@@ -12,18 +12,9 @@
 import matplotlib.pyplot as plt
 
 ''' 1. Classify day and night images
-    2. Visualize the misclassified imagess
-''' 
-
-
-'''Training and Testing Data
-The 92 day/night images are separated into training and testing datasets.
-
-52 of these images are training images, for you to use as you create a classifier.
-40 are test images, which will be used to test the accuracy of your classifier.
-First, we set some variables to keep track of some where our images are stored'''
-
+    2. Visualize the misclassified imagess ''' 
 
+'''First, we set some variables to keep track of some where our images are stored'''
 
 # Image data directories
 image_dir_training = "day_night_images/training/"

Day-Night-Classifier/DayNight_HSV_AvgBr.py

@@ -12,7 +12,7 @@
 import matplotlib.pyplot as plt
 
 ''' 
-1. Show HSV channels for one day image'
+1. Display HSV channels for one day image'
 2. Testing average brightness levels'''
 
 

Day-Night-Classifier/README.md

@@ -0,0 +1,33 @@
+
+## Training and Testing Data
+The 220 day/night images are separated into training and testing datasets.
+
+120 of these images are training images to create a classifier.
+100 are test images, which will be used to test the accuracy of classifier. 
+
+## Feature Extraction
+Create a feature that represents the brightness in an image. 
+We extract the average brightness using HSV colorspace.
+Specifically, we use the V channel (a measure of brightness),
+add up the pixel values in the V channel, then divide that sum 
+by the area of the image to get the average Value of the image.
+
+Find the average brightness using the V channel
+This function takes in a standardized RGB image and returns
+a feature (a single value) that represent the average level of 
+brightness in the image. We'll use this value to classify
+the image as day or night.
+
+## Day Night Image Classifier
+ * DayNight_HSV_AbgBr.py
+   1. Display HSV channels for one day image
+   2. Testing average brightness levels
+
+ * DayNight_Blue_Thresholds.py 
+   1. Day Night Threshold
+   2. Blue channel Threshold 
+
+ * DayNaight_Classifier.py
+   1. Classify day and night images
+   2. Visualize the misclassified imagess 
+