Object-Detection/shape_analyzer.py at master · DevRodox/Object-Detection · 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
import matplotlib.patches as mpatches
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from skimage.draw import disk, polygon, rectangle
from skimage import measure
class ShapeAnalyzer:
    def __init__(self, colored_images):
        self.colored_images = colored_images
    def calculate_properties(self, prop):
        properties = {
            'area': prop.area,
            'bounding_box': prop.bbox,
            'centroid': prop.centroid,
            'convex_area': prop.convex_area,
            'convex_image': prop.convex_image,
            'coordinates': prop.coords,
            'eccentricity': prop.eccentricity,
            'equivalent_diameter': prop.equivalent_diameter,
            'euler_number': prop.euler_number,
            'extent': prop.extent,
            'filled_area': prop.filled_area,
            'filled_image': prop.filled_image,
            'major_axis_length': prop.major_axis_length,
            'minor_axis_length': prop.minor_axis_length,
            'orientation': prop.orientation,
            'perimeter': prop.perimeter,
            'solidity': prop.solidity,
            'circularity': (4 * np.pi * prop.area) / (prop.perimeter ** 2) if prop.perimeter != 0 else 0,
            'convexity': prop.convex_area / prop.area if prop.area != 0 else 0,
        }

        if hasattr(prop, 'intensity_image'):
            properties['intensity'] = prop.mean_intensity if isinstance(prop.mean_intensity, (float, int)) else prop.median_intensity
        else:
            properties['intensity'] = None

        return properties
    def determine_shape(self, properties):
                num_vertices = len(properties['bounding_box'])

                aspect_ratio_range = (1.5, 2.5)

                minr, minc, maxr, maxc = properties['bounding_box']
                width = maxc - minc
                height = maxr - minr
                aspect_ratio = width / height

                if properties['circularity'] > 0.9 and properties['eccentricity'] < 0.1:
                    return 'Círculo'
                elif properties['convexity'] == 1 and properties['solidity'] < 0.9:
                    return 'Triángulo'
                elif aspect_ratio >= aspect_ratio_range[0] and aspect_ratio <= aspect_ratio_range[1]:
                    return 'Rectángulo'
                elif 0.9 <= aspect_ratio <= 1.1:
                    return 'Rectángulo'
                elif num_vertices == 3 or num_vertices == 4:
                    angles = [0, 0, 0, 0]
                    if width > 0 and height > 0:
                        angles[0] = np.arctan2(height, width) * 180 / np.pi
                        angles[1] = 90 - angles[0]
                        angles[2] = angles[0]
                        angles[3] = angles[1]

                    if sum(angles) == 180:
                        return 'Triángulo'
                    elif sum(angles) == 360:
                        return 'Rectángulo'
    def analyze_shapes(self, colored_image, ax):
                label_image = measure.label(np.any(colored_image > 0, axis=2))
                props = measure.regionprops(label_image)
                ax.imshow(colored_image)

                for prop in props:
                    shape_properties = self.calculate_properties(prop)
                    shape_type = self.determine_shape(shape_properties)
                    minr, minc, maxr, maxc = prop.bbox
                    rect = mpatches.Rectangle((minc, minr), maxc - minc, maxr - minr,
                                            fill=False, edgecolor='red', linewidth=2)
                    ax.add_patch(rect)
                    ax.text(minc, minr, shape_type, color='yellow', fontsize=12)
    def make_results_for_all_images(self):
        fig, axes = plt.subplots(1, len(self.colored_images), figsize=(30, 7))
        if len(self.colored_images) == 1:
            axes = [axes]

        category_titles = ['Circles', 'Triangles', 'Rectangles', 'Mezcla de los tres tipos']

        shape_counts = {
            'circles': {'Círculo': 0, 'Triángulo': 0, 'Rectángulo': 0},
            'triangles': {'Círculo': 0, 'Triángulo': 0, 'Rectángulo': 0},
            'rectangles': {'Círculo': 0, 'Triángulo': 0, 'Rectángulo': 0},
            'random_shapes': {'Círculo': 0, 'Triángulo': 0, 'Rectángulo': 0}
        }

        combined_data = []
        for i, colored_image in enumerate(self.colored_images):
            label_image = measure.label(np.any(colored_image > 0, axis=2))
            props = measure.regionprops(label_image)
            ax = axes[i] if len(self.colored_images) > 1 else axes
            ax.imshow(colored_image)
            ax.set_title(category_titles[i])

            current_shape_counts = {'Círculo': 0, 'Triángulo': 0, 'Rectángulo': 0}
            figures_data = []
            for prop in props:
                shape_properties = self.calculate_properties(prop)
                shape_type = self.determine_shape(shape_properties)
                current_shape_counts[shape_type] += 1

                minr, minc, maxr, maxc = prop.bbox
                rect = mpatches.Rectangle((minc, minr), maxc - minc, maxr - minr,
                                        fill=False, edgecolor='red', linewidth=2)
                ax.add_patch(rect)
                ax.text(minc, minr, shape_type, color='yellow', fontsize=12)

                figure_data = {'Category': category_titles[i]}
                figure_data.update(shape_properties)
                figures_data.append(figure_data)

            df_figures = pd.DataFrame(figures_data)
            combined_data.append(df_figures)

            category_names = ['circles', 'triangles', 'rectangles', 'random_shapes']
            if i < len(category_names):
                category_name = category_names[i]
                shape_counts[category_name] = current_shape_counts

        if combined_data:
            combined_df = pd.concat(combined_data, ignore_index=True)
            combined_df = combined_df[combined_df['Category'] != 'Mezcla de los tres tipos']
            combined_df = combined_df[['Category'] + [col for col in combined_df.columns if col != 'Category']]
            combined_df.to_csv('./Propiedades/Combinado.csv', index=False)
        else:
            print("No se pudo crear el archivo combinado porque no se encontraron archivos CSV para las categorías.")

        plt.tight_layout()
        plt.show()

        return shape_counts