Show1 (1).py

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import matplotlib.image as img
import pathlib
import warnings 
warnings.filterwarnings('ignore')
import cv2
import pickle
from numpy import linalg as LA

from tkinter import * 
from tkinter import messagebox
import os
import cv2
import numpy as np # linear algebra
import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)
pd.options.mode.chained_assignment = None  # default='warn'
from PIL import Image
from tqdm import tqdm
import pickle
#from skimage.feature import greycomatrix,greycoprops --OLD vERSION
from skimage.feature import graycomatrix, graycoprops
import skimage.feature as feature

import easygui

def fxnglcm(img,i,df):
    cv2.imwrite('0.png', img)
    img=cv2.imread('0.png',0)
    print(np.shape(img))
    graycom = feature.graycomatrix(img, [1], [0, np.pi/4, np.pi/2, 3*np.pi/4], levels=256)
    c = feature.graycoprops(graycom, 'contrast')
    d = feature.graycoprops(graycom, 'dissimilarity')
    h = feature.graycoprops(graycom, 'homogeneity')
    e = feature.graycoprops(graycom, 'energy')
    corr = feature.graycoprops(graycom, 'correlation')
    ASM = feature.graycoprops(graycom, 'ASM')
    c = np.squeeze(c)
    d = np.squeeze(d)
    h = np.squeeze(h)
    e = np.squeeze(e)
    corr = np.squeeze(corr)
    asm = np.squeeze(ASM)
    
    df["GLCM-Contrast-1"][i] = c[0]
    df["GLCM-Contrast-2"][i] = c[1]
    df["GLCM-Contrast-3"][i] = c[2]
    df["GLCM-Contrast-4"][i] = c[3]

    df["GLCM-Dissimilarity-1"][i] =d[0]
    df["GLCM-Dissimilarity-2"][i] =d[1]
    df["GLCM-Dissimilarity-3"][i] =d[2]
    df["GLCM-Dissimilarity-4"][i] =d[3]
    
    df["GLCM-Homogeneity-1"][i] =h[0]
    df["GLCM-Homogeneity-2"][i]=h[1]
    df["GLCM-Homogeneity-3"][i] =h[2]
    df["GLCM-Homogeneity-4"][i] =h[3]
    
    df["GLCM-Energy-1"][i] =e[0]
    df["GLCM-Energy-2"][i] =e[1]
    df["GLCM-Energy-3"][i] =e[2]
    df["GLCM-Energy-4"][i] =e[3]

    df["GLCM-Correlation-1"][i] =corr[0]
    df["GLCM-Correlation-2"][i] =corr[1]
    df["GLCM-Correlation-3"][i] =corr[2]
    df["GLCM-Correlation-4"][i] =corr[3]

    df["GLCM-ASM-1"][i] =asm[0]
    df["GLCM-ASM-2"][i] =asm[1]
    df["GLCM-ASM-3"][i] =asm[2]
    df["GLCM-ASM-4"][i] =asm[3]
    
    return df

   
def glcmfiller(imgarr,i,gry,df):
    if imgarr.shape !=(224, 224):
        og = cv2.cvtColor(imgarr,cv2.COLOR_BGR2GRAY)
        fxnglcm(og,i,df)
        df["gry"][i] = og
    else:
        fxnglcm(imgarr,i,df)
        df["gry"][i] = imgarr
    return



file= open("model.h5",'rb')
TRNFV = pickle.load(file)
file.close()



def Calc_Wt(TRR,TST):
    WTRN = TRR
    R, C = np.shape(WTRN)
    M = []
    ERR = []
    WTST = TST
    R, C = np.shape(WTRN)
    print('RC',np.shape(WTRN))
    for i in range(0, C):
        RR = WTRN[:,i]
        Temp = np.subtract(WTST, RR)
        ERR = LA.norm(Temp)
        M.append(ERR)
    ind = np.argmin(M);
    return ind



from tkinter import filedialog
# VIDEO READ
fnam= filedialog.askopenfilename(filetypes = (("BROWSE ", "*.png"), ("All files", "*")))


# -----------------------------------------------
#               MAIN CODE
# -----------------------------------------------
# READ IMAGE
img=cv2.imread(fnam)
a=img
cv2.imshow('INPUT IMAGE',img)
cv2.waitKey(100)
cv2.destroyAllWindows()

# RESIZING
img = cv2.resize(img,(256,256),3)
cv2.imshow('RESIZED IMAGE',img)
cv2.waitKey(100)
cv2.destroyAllWindows()

# MEDIAN FILTERED
img1 = cv2.medianBlur(img,5)
cv2.imshow('MEDIAN IMAGE',img1)
cv2.waitKey(100)
cv2.destroyAllWindows()


file= open("model.h5",'rb')
CNN = pickle.load(file)
file.close()

#CNN=np.transpose(CNN)


img= cv2.cvtColor(a,cv2.COLOR_BGR2GRAY)
graycom = feature.graycomatrix(img, [1], [0, np.pi/4, np.pi/2, 3*np.pi/4], levels=256)
c = feature.graycoprops(graycom, 'contrast')
d = feature.graycoprops(graycom, 'dissimilarity')
h = feature.graycoprops(graycom, 'homogeneity')
e = feature.graycoprops(graycom, 'energy')
corr = feature.graycoprops(graycom, 'correlation')
ASM = feature.graycoprops(graycom, 'ASM')
c = np.squeeze(c)
d = np.squeeze(d)
h = np.squeeze(h)
e = np.squeeze(e)
corr = np.squeeze(corr)
asm = np.squeeze(ASM)
FV=[c,d,h,e,corr,asm]
FV=np.ravel(FV)

print('index',np.shape(FV),np.shape(CNN))
n=Calc_Wt(CNN,FV)
print('index',n)



if n<=50:
    Dis='NO'
else:
    Dis='YES'
print('-------------------------------------------------------\n')
print('                     ',Dis,'                         \n')
print('-------------------------------------------------------\n')


if n<=50:
    Dis='NORMAL'
    easygui.msgbox(" RESULT --> NORMAL... !", title="simple gui")
else:
    Dis='TB PRESENT'
    easygui.msgbox(" RESULT --> TB PRESENT !", title="simple gui")
print('-------------------------------------------------------\n')
print('                     ',Dis,'                         \n')
print('-------------------------------------------------------\n')