SnowCoverageEstimation-/tools.py at main · kadavparth/SnowCoverageEstimation- · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
from lib2to3.pgen2.token import RPAR
import os,sys
from webbrowser import get
import pandas as pd
import cv2
import json
import numpy as np
import time
from sklearn.preprocessing import normalize
import json
import csv

#------------------------------------------------------------
# Preprocessing and Image transformation functions
#------------------------------------------------------------

def read_json(filename):
    '''
    Description
    ----------
    Read parameters from json and return dict.

    Parameters
    ----------
    filename  :  str
        Name of json file to extract values from.

    Return Type  -->  dict
    '''
    PARAM = {}
    try:
        with open(filename, "r") as f:
            data = json.loads(f.read())
            for key, value in data.items():
                PARAM[key] = value
    except FileNotFoundError:
        #print("{} --> DIDN'T FIND {}!".format('FileNotFoundError', filename))
        PARAM = None
    return PARAM

def adjust_with_max(pt, max_x, max_y):
    if pt[0] > max_x:
        new_x = max_x
    else:
        new_x = pt[0]

    if pt[1] > max_y:
        new_y = max_y
    else:
        new_y = pt[1]

    return (new_x, new_y)

def draw_rois(img, ROI_PARAMS):

    lane_mask = np.zeros((720,1280),dtype='uint8')
    road_mask = np.zeros((720,1280),dtype='uint8')

    h = 720
    w = 1280

    # ##################### LANE ROI ####################
    # Calculate points based on ROI_PARAMS
    lane_bl = adjust_with_max((ROI_PARAMS["lane_roi"]["roi_center_x"] - 0.5*ROI_PARAMS["lane_roi"]["bottom_width"], ROI_PARAMS["lane_roi"]["bottom_y"]), w, h)
    lane_tl = adjust_with_max((ROI_PARAMS["lane_roi"]["roi_center_x"] - 0.5*ROI_PARAMS["lane_roi"]["top_width"], ROI_PARAMS["lane_roi"]["top_y"]), w, h)
    lane_br = adjust_with_max((ROI_PARAMS["lane_roi"]["roi_center_x"] + 0.5*ROI_PARAMS["lane_roi"]["bottom_width"], ROI_PARAMS["lane_roi"]["bottom_y"]), w, h)
    lane_tr = adjust_with_max((ROI_PARAMS["lane_roi"]["roi_center_x"] + 0.5*ROI_PARAMS["lane_roi"]["top_width"], ROI_PARAMS["lane_roi"]["top_y"]), w, h)

    # filter points for max x and y

    lane_pts = np.array((lane_bl, lane_tl, lane_tr, lane_br), np.int32)
    lane_mask = cv2.fillPoly(lane_mask, [lane_pts], color=(255,255,255))

    # UNCOMMENT BELOW LINE TO SHOW LANE ROI
    # out_img = cv2.polylines(out_img, [lane_pts], True, (0, 0, 255), 2)

    # ##################### ROAD ROI ####################
    # Calculate points based on ROI_PARAMS
    road_bl = adjust_with_max((ROI_PARAMS["road_roi"]["roi_center_x"] - 0.5*ROI_PARAMS["road_roi"]["bottom_width"] - ROI_PARAMS["road_roi"]["diff"], ROI_PARAMS["road_roi"]["bottom_y"]), w, h)
    road_tl = adjust_with_max((ROI_PARAMS["road_roi"]["roi_center_x"] - 0.5*ROI_PARAMS["road_roi"]["top_width"] + ROI_PARAMS["road_roi"]["diff"], ROI_PARAMS["road_roi"]["top_y"]), w, h)
    road_br = adjust_with_max((ROI_PARAMS["road_roi"]["roi_center_x"] + 0.5*ROI_PARAMS["road_roi"]["bottom_width"] - ROI_PARAMS["road_roi"]["diff"], ROI_PARAMS["road_roi"]["bottom_y"]), w, h)
    road_tr = adjust_with_max((ROI_PARAMS["road_roi"]["roi_center_x"] + 0.5*ROI_PARAMS["road_roi"]["top_width"] + ROI_PARAMS["road_roi"]["diff"], ROI_PARAMS["road_roi"]["top_y"]), w, h)

    # filter points for max x and y

    road_pts = np.array((road_bl, road_tl, road_tr, road_br), np.int32)
    road_mask = cv2.fillPoly(road_mask, [road_pts], color=(255,255,255))
    road_mask = cv2.resize(road_mask, (256,256))
    # out_img = cv2.polylines(out_img, [road_pts], True, (255, 0, 0), 2)
    return lane_mask, road_mask

def get_roi_pixels(raw_image_resize,z):
    v = cv2.bitwise_and(raw_image_resize, raw_image_resize, mask=z)
    pixel_locs = np.argwhere(z == 255)
    roi_pixels = np.ndarray((len(pixel_locs), 3))
    for i, loc in enumerate(pixel_locs):
        roi_pixels[i] = raw_image_resize[loc[0], loc[1]]
    # reshape the image to a 2D array of pixels and 1 color values (Gray)
    # Reshaping to 1D vector
    pixel_values = roi_pixels.reshape((-1, 1))
    pixel_values = np.float32(pixel_values)

    return pixel_values, pixel_locs

def get_avg_ch_vals(img, mask, values="bgr"):
    # get location of mask pixels
    pixel_locs = np.argwhere(mask == 255)
    b_vals = []
    g_vals = []
    r_vals = []
    for pixel in pixel_locs:
        cube = img[pixel[0], pixel[1]]
        b_vals.append(cube[0])
        g_vals.append(cube[1])
        r_vals.append(cube[2])

    return np.average(b_vals), np.average(g_vals), np.average(r_vals)

def get_std_ch_vals(img, mask, values = "bgr"):
    pixel_locs = np.argwhere(mask == 255)
    b_vals = []
    g_vals = []
    r_vals = []
    for pixel in pixel_locs:
        cube = img[pixel[0], pixel[1]]
        b_vals.append(cube[0])
        g_vals.append(cube[1])
        r_vals.append(cube[2])

    return np.std(b_vals), np.average(g_vals), np.average(r_vals)

# -------------------------------------------------------------
# Data preparation and feature extraction functions
# -------------------------------------------------------------
def preprocessImage(image, roi_mask):
    image = cv2.resize(image, (256,256))
    masked_image = cv2.bitwise_and(image, image, mask=roi_mask)
    return masked_image

# def preprocessLabel(label, roi_mask, size):
#     label = cv2.resize(label, size)
#     masked_label = cv2.bitwise_and(label, label, mask=roi_mask)
#     _, masked_threshed_label = cv2.threshold(masked_label, 50, 255, cv2.THRESH_BINARY)
#     return masked_threshed_label

def getFeatureVector(df_val,image,image_256,ROI_PARAMS,features=['rgb-mean', 'rgb-std', 'rgb-mean-std', 'rgb-mean-vis', 'rgb-std-vis', 'rgb-mean-std-vis']):
    feat_list = []
    lane_mask, road_mask = draw_rois(image, ROI_PARAMS)
    road_vals_mean = get_avg_ch_vals(image_256, road_mask)
    road_vals_std = get_std_ch_vals(image_256, road_mask)
    for feature in features:
        if feature == 'rgb-mean':
            feat_list.append([road_vals_mean[2], road_vals_mean[1], road_vals_mean[0]])
        elif feature == 'rgb-std':
            feat_list.append([road_vals_std[2], road_vals_std[1], road_vals_std[0]])
        elif feature == 'rgb-mean-std':
            feat_list.append([road_vals_mean[2], road_vals_mean[1], road_vals_mean[0], road_vals_std[2], road_vals_std[1], road_vals_std[0]])
        elif feature == 'rgb-mean-vis':
            feat_list.append([road_vals_mean[2], road_vals_mean[1], road_vals_mean[0], df_val])
        elif feature == 'rgb-std-vis':
            feat_list.append([road_vals_std[2], road_vals_std[1], road_vals_std[0], df_val])
        elif feature == 'rgb-mean-std-vis':
            feat_list.append([road_vals_mean[2], road_vals_mean[1], road_vals_mean[0], road_vals_std[2], road_vals_std[1], road_vals_std[0], df_val])
    file = open('/home/parth/AIM2/snow_coverage_estimation/1_get_feature_and_label_sets/all.csv', 'a+', newline='')
    with file:
        write = csv.writer(file)
        write.writerows(feat_list)
    return feat_list