EDA fruits360

In [1]:

!pip install gdown

!pip install gdown

Requirement already satisfied: gdown in /usr/local/lib/python3.10/dist-packages (5.2.0)
Requirement already satisfied: beautifulsoup4 in /usr/local/lib/python3.10/dist-packages (from gdown) (4.12.3)
Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from gdown) (3.16.1)
Requirement already satisfied: requests[socks] in /usr/local/lib/python3.10/dist-packages (from gdown) (2.32.3)
Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from gdown) (4.66.5)
Requirement already satisfied: soupsieve>1.2 in /usr/local/lib/python3.10/dist-packages (from beautifulsoup4->gdown) (2.6)
Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.10/dist-packages (from requests[socks]->gdown) (3.4.0)
Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests[socks]->gdown) (3.10)
Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests[socks]->gdown) (2.2.3)
Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests[socks]->gdown) (2024.8.30)
Requirement already satisfied: PySocks!=1.5.7,>=1.5.6 in /usr/local/lib/python3.10/dist-packages (from requests[socks]->gdown) (1.7.1)

In [2]:

import gdown
import os
import zipfile
import numpy as np
import matplotlib.pyplot as plt
from PIL import Image, ImageEnhance
from skimage import feature
import cv2

import gdown import os import zipfile import numpy as np import matplotlib.pyplot as plt from PIL import Image, ImageEnhance from skimage import feature import cv2

In [3]:

file_id = "18fRamctgsdMh0E0DgxYX_ZdQs2e_Qq73"
gdown.download(f"https://drive.google.com/uc?id={file_id}", "fruits360.zip", quiet=False)

file_id = "18fRamctgsdMh0E0DgxYX_ZdQs2e_Qq73" gdown.download(f"https://drive.google.com/uc?id={file_id}", "fruits360.zip", quiet=False)

Downloading...
From (original): https://drive.google.com/uc?id=18fRamctgsdMh0E0DgxYX_ZdQs2e_Qq73
From (redirected): https://drive.google.com/uc?id=18fRamctgsdMh0E0DgxYX_ZdQs2e_Qq73&confirm=t&uuid=5dd64a6c-8f08-43bb-baf6-4d9c624415a6
To: /content/fruits360.zip
100%|██████████| 1.01G/1.01G [00:25<00:00, 40.5MB/s]

Out[3]:

'fruits360.zip'

In [4]:

zip_file_name = "fruits360.zip"
os.makedirs("dataset", exist_ok=True)

with zipfile.ZipFile(zip_file_name, "r") as zip_ref:
    zip_ref.extractall("dataset")

zip_file_name = "fruits360.zip" os.makedirs("dataset", exist_ok=True) with zipfile.ZipFile(zip_file_name, "r") as zip_ref: zip_ref.extractall("dataset")

In [5]:

# DATASET_DIR = './dataset/fruits360/fruits-360_dataset_100x100/fruits-360/'
DATASET_DIR = "./dataset/fruits360/fruits-360_dataset_original-size/fruits-360-original-size/"
SET_PATHS_IMAGES = set()
WIDTHS = []
HEIGHTS = []

# DATASET_DIR = './dataset/fruits360/fruits-360_dataset_100x100/fruits-360/' DATASET_DIR = "./dataset/fruits360/fruits-360_dataset_original-size/fruits-360-original-size/" SET_PATHS_IMAGES = set() WIDTHS = [] HEIGHTS = []

In [6]:

def init_set_paths_images():
    for root, dirs, files in os.walk(DATASET_DIR):
        for file in files:
            if file.endswith((".png", ".jpg", ".jpeg")):
                file_path = os.path.join(root, file)
                SET_PATHS_IMAGES.add(file_path)

def init_set_paths_images(): for root, dirs, files in os.walk(DATASET_DIR): for file in files: if file.endswith((".png", ".jpg", ".jpeg")): file_path = os.path.join(root, file) SET_PATHS_IMAGES.add(file_path)

In [7]:

def get_list_size_img() -> list[int]:
    widths = []
    heights = []
    if len(SET_PATHS_IMAGES) == 0:
        return
    for file_path in SET_PATHS_IMAGES:
        with Image.open(file_path) as img:
            widths.append(img.width)
            heights.append(img.height)
    return widths, heights

def get_list_size_img() -> list[int]: widths = [] heights = [] if len(SET_PATHS_IMAGES) == 0: return for file_path in SET_PATHS_IMAGES: with Image.open(file_path) as img: widths.append(img.width) heights.append(img.height) return widths, heights

In [8]:

def get_mean_std_chanels_img() -> dict[str, float]:
    sum_mean_r = 0
    sum_mean_g = 0
    sum_mean_b = 0
    sum_r_std = 0
    sum_g_std = 0
    sum_b_std = 0
    image_count = 0

    if len(SET_PATHS_IMAGES) == 0:
        return
    for file_path in SET_PATHS_IMAGES:
        with Image.open(file_path) as img:
            img = img.convert("RGB")
            np_img = np.array(img).astype(np.float64)
            r = np_img[:, :, 0]
            g = np_img[:, :, 1]
            b = np_img[:, :, 2]

            sum_mean_r += np.mean(r)
            sum_mean_g += np.mean(g)
            sum_mean_b += np.mean(b)

            sum_r_std += np.std(r)
            sum_g_std += np.std(g)
            sum_b_std += np.std(b)

            image_count += 1

    return {
        "mean_r": sum_mean_r / image_count,
        "mean_g": sum_mean_g / image_count,
        "mean_b": sum_mean_b / image_count,
        "std_r": sum_r_std / image_count,
        "std_g": sum_g_std / image_count,
        "std_b": sum_b_std / image_count,
    }

def get_mean_std_chanels_img() -> dict[str, float]: sum_mean_r = 0 sum_mean_g = 0 sum_mean_b = 0 sum_r_std = 0 sum_g_std = 0 sum_b_std = 0 image_count = 0 if len(SET_PATHS_IMAGES) == 0: return for file_path in SET_PATHS_IMAGES: with Image.open(file_path) as img: img = img.convert("RGB") np_img = np.array(img).astype(np.float64) r = np_img[:, :, 0] g = np_img[:, :, 1] b = np_img[:, :, 2] sum_mean_r += np.mean(r) sum_mean_g += np.mean(g) sum_mean_b += np.mean(b) sum_r_std += np.std(r) sum_g_std += np.std(g) sum_b_std += np.std(b) image_count += 1 return { "mean_r": sum_mean_r / image_count, "mean_g": sum_mean_g / image_count, "mean_b": sum_mean_b / image_count, "std_r": sum_r_std / image_count, "std_g": sum_g_std / image_count, "std_b": sum_b_std / image_count, }

In [9]:

def scatter_size_images():
    plt.figure(figsize=(10, 6))
    plt.scatter(WIDTHS, HEIGHTS, color="blue", alpha=0.5)
    plt.title("Распределение размеров изображений (ширина vs высота)")
    plt.xlabel("Ширина")
    plt.ylabel("Высота")
    plt.grid()

    plt.tight_layout()
    plt.show()

def scatter_size_images(): plt.figure(figsize=(10, 6)) plt.scatter(WIDTHS, HEIGHTS, color="blue", alpha=0.5) plt.title("Распределение размеров изображений (ширина vs высота)") plt.xlabel("Ширина") plt.ylabel("Высота") plt.grid() plt.tight_layout() plt.show()

In [10]:

def bar(classes, counts):
    plt.figure(figsize=(35, 8))
    plt.bar(classes, counts, color="#D2B48C")
    y_pos = np.arange(len(classes))
    plt.xticks(y_pos, classes, rotation=90, ha="right", fontsize=10)
    plt.subplots_adjust(bottom=0.2)
    plt.ylabel("Количество изображений")
    plt.title("Распределение классов изображений")
    plt.show()

def bar(classes, counts): plt.figure(figsize=(35, 8)) plt.bar(classes, counts, color="#D2B48C") y_pos = np.arange(len(classes)) plt.xticks(y_pos, classes, rotation=90, ha="right", fontsize=10) plt.subplots_adjust(bottom=0.2) plt.ylabel("Количество изображений") plt.title("Распределение классов изображений") plt.show()

In [11]:

def resize_img(new_size: tuple):
    if len(SET_PATHS_IMAGES) == 0:
        return
    for file_path in SET_PATHS_IMAGES:
        with Image.open(file_path) as img:
            if tuple(img.size) != new_size:
                img_resized = img.resize(new_size)
                img_resized.save(file_path)

def resize_img(new_size: tuple): if len(SET_PATHS_IMAGES) == 0: return for file_path in SET_PATHS_IMAGES: with Image.open(file_path) as img: if tuple(img.size) != new_size: img_resized = img.resize(new_size) img_resized.save(file_path)

In [12]:

def rotate_img(img, angle) -> Image:
    img_rotated = img.rotate(angle)
    return img_rotated

def rotate_img(img, angle) -> Image: img_rotated = img.rotate(angle) return img_rotated

In [13]:

def brightness_img(img, brightness_coeff) -> Image:
    enhancer = ImageEnhance.Brightness(img)
    img_brightened = enhancer.enhance(brightness_coeff)
    return img_brightened

def brightness_img(img, brightness_coeff) -> Image: enhancer = ImageEnhance.Brightness(img) img_brightened = enhancer.enhance(brightness_coeff) return img_brightened

In [14]:

def contrasted_img(img, contrast_coeff) -> Image:
    enhancer = ImageEnhance.Contrast(img)
    img_contrasted = enhancer.enhance(contrast_coeff)
    return img_contrasted

def contrasted_img(img, contrast_coeff) -> Image: enhancer = ImageEnhance.Contrast(img) img_contrasted = enhancer.enhance(contrast_coeff) return img_contrasted

In [15]:

def get_HOG_img(img, orientations: int = 9, pixels_per_cell: tuple = (8, 8), cells_per_block: tuple = (2, 2)) -> Image:
    hog_image = feature.hog(
        img,
        orientations=orientations,
        pixels_per_cell=pixels_per_cell,
        cells_per_block=cells_per_block,
        visualize=True,
    )
    return hog_image

def get_HOG_img(img, orientations: int = 9, pixels_per_cell: tuple = (8, 8), cells_per_block: tuple = (2, 2)) -> Image: hog_image = feature.hog( img, orientations=orientations, pixels_per_cell=pixels_per_cell, cells_per_block=cells_per_block, visualize=True, ) return hog_image

In [16]:

def get_SIFT_img(img) -> Image:
    sift = cv2.SIFT_create()
    keypoints, descriptors = sift.detectAndCompute(img, None)
    output_image = cv2.drawKeypoints(img, keypoints, None)

    return output_image

def get_SIFT_img(img) -> Image: sift = cv2.SIFT_create() keypoints, descriptors = sift.detectAndCompute(img, None) output_image = cv2.drawKeypoints(img, keypoints, None) return output_image

In [17]:

init_set_paths_images()

init_set_paths_images()

In [18]:

num_images = len(SET_PATHS_IMAGES)
print(f"Количество изображений: {num_images}")

num_images = len(SET_PATHS_IMAGES) print(f"Количество изображений: {num_images}")

Количество изображений: 12455

In [19]:

WIDTHS, HEIGHTS = get_list_size_img()
size = np.array(list(zip(WIDTHS, HEIGHTS)))
mean_width = int(np.mean(WIDTHS))
mean_height = int(np.mean(HEIGHTS))
min_width = np.min(WIDTHS)
min_height = np.min(HEIGHTS)
max_width = np.max(WIDTHS)
max_height = np.max(HEIGHTS)


print(f"средняя ширина: {mean_width}")
print(f"средняя высота: {mean_height}")
print(f"минимальная ширина: {min_width}")
print(f"минимальная высота: {min_height}")
print(f"максимальная ширина: {max_width}")
print(f"максимальная высота: {max_height}")

WIDTHS, HEIGHTS = get_list_size_img() size = np.array(list(zip(WIDTHS, HEIGHTS))) mean_width = int(np.mean(WIDTHS)) mean_height = int(np.mean(HEIGHTS)) min_width = np.min(WIDTHS) min_height = np.min(HEIGHTS) max_width = np.max(WIDTHS) max_height = np.max(HEIGHTS) print(f"средняя ширина: {mean_width}") print(f"средняя высота: {mean_height}") print(f"минимальная ширина: {min_width}") print(f"минимальная высота: {min_height}") print(f"максимальная ширина: {max_width}") print(f"максимальная высота: {max_height}")

средняя ширина: 453
средняя высота: 531
минимальная ширина: 96
минимальная высота: 250
максимальная ширина: 950
максимальная высота: 1061

In [20]:

dict_mean_std_chanels_img = get_mean_std_chanels_img()
mean_r = round(dict_mean_std_chanels_img["mean_r"], 2)
mean_g = round(dict_mean_std_chanels_img["mean_g"], 2)
mean_b = round(dict_mean_std_chanels_img["mean_b"], 2)
std_r = round(dict_mean_std_chanels_img["std_r"], 2)
std_g = round(dict_mean_std_chanels_img["std_g"], 2)
std_b = round(dict_mean_std_chanels_img["std_b"], 2)
print(f"Средние значения: R = {mean_r}, G = {mean_g}, B = {mean_b}")
print(f"Стандартные отклонения: R = {std_r}, G = {std_g}, B = {std_b}")

dict_mean_std_chanels_img = get_mean_std_chanels_img() mean_r = round(dict_mean_std_chanels_img["mean_r"], 2) mean_g = round(dict_mean_std_chanels_img["mean_g"], 2) mean_b = round(dict_mean_std_chanels_img["mean_b"], 2) std_r = round(dict_mean_std_chanels_img["std_r"], 2) std_g = round(dict_mean_std_chanels_img["std_g"], 2) std_b = round(dict_mean_std_chanels_img["std_b"], 2) print(f"Средние значения: R = {mean_r}, G = {mean_g}, B = {mean_b}") print(f"Стандартные отклонения: R = {std_r}, G = {std_g}, B = {std_b}")

Средние значения: R = 158.03, G = 132.29, B = 102.46
Стандартные отклонения: R = 66.03, G = 75.76, B = 85.91

In [21]:

classes_patch = DATASET_DIR + "Training/"
classes = os.listdir(classes_patch)
counts = []
for img in classes:
    counts.append(len(os.listdir(os.path.join(classes_patch, img))))

classes_patch = DATASET_DIR + "Training/" classes = os.listdir(classes_patch) counts = [] for img in classes: counts.append(len(os.listdir(os.path.join(classes_patch, img))))

In [22]:

bar(classes, counts)

bar(classes, counts)

In [23]:

scatter_size_images()

scatter_size_images()