from PIL import Image
import numpy as np
from sklearn.ensemble import IsolationForest
from collections import Counter
# 判斷rgb相似度用，可以小一些
threhold=15
threhold_border=15
# 采樣半徑，可以設(shè)的小一些，可以根據(jù)邊框的粗細(xì)來設(shè)定
redius=20
redius_border=20
# 對(duì)邊框進(jìn)行處理需要調(diào)整的參數(shù)
step_add=37
step_max=2000
step=0
all_outliers=[]
top_outlier_num=10
# 創(chuàng)建Isolation Forest模型
model = IsolationForest(contamination=0.05)  # 設(shè)置異常點(diǎn)的比例

#判斷顏色是否相似
def my_similar(a,b):
    if sum([abs(x-y) for (x,y) in zip(list(a),list(b))])/3<threhold:
        return True
    return False
# 判斷顏色是否相似
def my_similar_border(a,b_s):
    for b in b_s:
        if sum([abs(x-y) for (x,y) in zip(list(a),list(b)[0])])/3<threhold_border:
            return True
    return False
# 獲取異常值，用于處理邊框的邊緣Q
def get_outlier(tmp,model):
    normal_point=[]
    unnormal_point=[]
    # 將數(shù)據(jù)傳遞給模型進(jìn)行訓(xùn)練
    model.fit(tmp)
    # 獲取每個(gè)數(shù)據(jù)點(diǎn)的預(yù)測(cè)標(biāo)簽，-1 表示異常點(diǎn)，1 表示正常點(diǎn)
    labels = model.predict(tmp)
    # 打印每個(gè)數(shù)據(jù)點(diǎn)的標(biāo)簽
    for i, label in enumerate(labels):
        if label == -1:
            unnormal_point.append(tmp[i])
        if label==1:
            normal_point.append(tmp[i])
    return normal_point,unnormal_point
# 在水平或處置方向上獲取鄰居節(jié)點(diǎn)，以便進(jìn)行顏色替換
def get_right_neighborhood(target_color,neighborhood_x,neighborhood_y):
    diff_x=0
    diff_y=0
    for neighborhood in neighborhood_x:
        diff_x+=abs(sum([x-y for x,y in zip(list(neighborhood),list(target_color))])/3)
    for neighborhood in neighborhood_y:
        diff_y+=abs(sum([x-y for x,y in zip(list(neighborhood),list(target_color))])/3)
    if diff_x>diff_y:
        return neighborhood_x,1
    return neighborhood_y,2

########################################################################################
############################開始對(duì)區(qū)域P和點(diǎn)A進(jìn)行人工指定################################
################################對(duì)應(yīng)“思路”第1部分#####################################
########################################################################################
# 利用手動(dòng)或利用labelimg等標(biāo)注工具，
# 確定“思路”第1部分中的區(qū)域“P”
#人工框選畫框的大致位置，需要包裹住畫框
x_1=1427
y_1=723
x_2=2061
y_2=1363
# 利用手動(dòng)或利用labelimg等標(biāo)注工具，
# 確定“思路”第1部分中的點(diǎn)“A”
x = 1495 # 目標(biāo)像素點(diǎn)的 x 坐標(biāo)
y = 1294 # 目標(biāo)像素點(diǎn)的 y 坐標(biāo)
# 主要函數(shù)
def replace_color_around_point(image_path, x, y, radius=redius):
    # 采樣步數(shù)
    step=0
    # 打開圖像
    image = Image.open(image_path)
    pixels = image.load()
    # 獲取圖像的寬度和高度
    width, height = image.size
    ########################################################################################
    ############################開始對(duì)邊框的內(nèi)部節(jié)點(diǎn)進(jìn)行處理################################
    ################################對(duì)應(yīng)“思路”第2部分#####################################
    ########################################################################################
    # 獲取目標(biāo)像素點(diǎn)的顏色
    target_color = pixels[x, y]
    # 循環(huán)遍歷圖像中的每個(gè)像素點(diǎn)
    for i in range(x_1,x_2):
        for j in range(y_1,y_2):
            # 如果像素顏色與目標(biāo)顏色相同，則替換為周圍區(qū)域顏色的平均值
            if my_similar(pixels[i, j],target_color):
                # 計(jì)算周圍水平和垂直區(qū)域的顏色平均值
                neighborhood_x = []
                neighborhood_y=[]
                for m in range(i - redius, i + radius + 1):
                        if 0 <= m < width:
                            neighborhood_x.append(pixels[m, j])
                for n in range(j - redius, j + radius + 1):
                        if 0 <= n < height:
                            neighborhood_y.append(pixels[i, n])
                neighborhood,direction=get_right_neighborhood(target_color,neighborhood_x,neighborhood_y)
                neighborhood=[n for n in neighborhood if not my_similar(n,target_color)]
                neighborhood = np.array(neighborhood)
                average_color=tuple(np.mean(neighborhood, axis=0, dtype=int))
                average_color_part_1 = tuple(np.mean(neighborhood[0:int(len(neighborhood)/2)], axis=0, dtype=int))
                average_color_part_2 = tuple(np.mean(neighborhood[int(len(neighborhood)/2)+1:], axis=0, dtype=int))
                # 替換像素顏色
                pixels[i, j] = average_color
                # 如果是水平方向
                if direction==1:
                    for m in range(i - int(redius_border/3), i):
                        if 0 <= m < width:
                                pixels[m,j]=average_color_part_1
                    for m in range(i, i + int(redius_border/3) + 1):
                        if 0 <= m < width:
                                pixels[m,j]=average_color_part_2
                # 如果是垂直方向
                if direction==2:            
                    for n in range(j - int(redius_border/3), j):
                        if 0 <= n < height:
                                pixels[i,n]=average_color_part_1
                    for n in range(j ,j + int(redius_border/3) + 1):
                        if 0 <= n < height:
                                pixels[i,n]=average_color_part_2
                step+=1
                if step%step_add==0 and step<step_max:
                    normal_point,unnormal_point=get_outlier(neighborhood,model)
                    unnormal_point=[tuple(x) for x in unnormal_point]
                    all_outliers.extend(unnormal_point)
    ########################################################################################
    ############################開始對(duì)邊框的邊緣節(jié)點(diǎn)進(jìn)行處理################################
    ################################對(duì)應(yīng)“思路”第2部分#####################################
    ########################################################################################
    # 使用Counter來統(tǒng)計(jì)三元組的出現(xiàn)次數(shù)
    all_outlier_counts = Counter(all_outliers)
    # 獲取出現(xiàn)次數(shù)最多的十個(gè)三元組，用來去邊框
    top_outliers = all_outlier_counts.most_common(top_outlier_num)
    # 循環(huán)遍歷圖像中的每個(gè)像素點(diǎn)
    for i in range(x_1,x_2):
        for j in range(y_1,y_2):
            # 如果該點(diǎn)和異常點(diǎn)中的顏色相似，就進(jìn)行替換
            if my_similar_border(pixels[i, j] ,top_outliers):
                # 計(jì)算周圍水平和垂直區(qū)域的顏色平均值
                neighborhood_x = []
                neighborhood_y=[]
                for m in range(i - int(radius/3), i + int(radius/3) + 1):
                        if 0 <= m < width:
                            neighborhood_x.append(pixels[m, j])
                for n in range(j - int(radius/3), j + int(radius/3) + 1):
                        if 0 <= n < height:
                            neighborhood_y.append(pixels[i, n])
                neighborhood,direction=get_right_neighborhood(target_color,neighborhood_x,neighborhood_y)
                neighborhood=[n for n in neighborhood if not n in top_outliers]
                neighborhood = np.array(neighborhood)
                average_color=tuple(np.mean(neighborhood, axis=0, dtype=int))
                # 替換像素顏色
                pixels[i, j] = average_color
    # 保存修改后的圖像
    image.save(r"../data/bbb.jpg")

image_path = r"../data/aaa.jpg"  # 替換為您的輸入圖像路徑
replace_color_around_point(image_path, x, y)