3D_Build.py

#!/usr/bin/env python
# coding: utf-8

# In[ ]:


import copy
import json
import os
import shutil
import cv2
import time
import numpy as np
from matplotlib import pyplot as plt
from IPython.display import Image
import import_ipynb
from rotate_translate import node
from parent_child_tree import createtree
from multipleoperation import *
from subprocess import run
import open3d as o3d
import time
from pyntcloud import PyntCloud


# Below Code generates 2 Contours for a given shape. One 'Outer Contour' and other 'Inner Contour' for a 2 shape .
# 

# In[ ]:


print("Please enter the image view file with extension")
s=input()
image=cv2.imread(s)
imgray=cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
ret, thresh = cv2.threshold(imgray,127,255,0,cv2.THRESH_BINARY_INV)
contours, hierarchy = cv2.findContours(thresh,cv2.RETR_CCOMP,cv2.CHAIN_APPROX_NONE)
hierarchy,contours = zip(*sorted(zip(hierarchy[0],contours),key = lambda x: cv2.contourArea(x[1]),reverse=True))
print ("no, of contours = " + str(len(contours)-1))
cv2.drawContours(image, contours, 1, (0, 255, 0), 2)
cv2.drawContours(image, contours, 2, (0, 0 , 255), 2)
cv2.imshow('IMAGE', image)
cv2.imshow('Image Gray', imgray)
cv2.waitKey(0)


# In[ ]:


print("Please enter the image view file with extension")
s=input()
image=cv2.imread(s)
imgray=cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
ret, thresh = cv2.threshold(imgray,127,255,0,cv2.THRESH_BINARY_INV)
contours, hierarchy = cv2.findContours(thresh,cv2.RETR_CCOMP,cv2.CHAIN_APPROX_NONE)
hierarchy,contours = zip(*sorted(zip(hierarchy[0],contours),key = lambda x: cv2.contourArea(x[1]),reverse=True))


# Below Function 'detect()' will determine the shape of a given 2D image on the basis of number of edges in the given Shape . 

# Allowed Shapes for a 2d Image are : Triangle , Square , Rectangle , Pentagon , Hexagon , Circle 
# 

# In[ ]:


def detect(c):
    perimeter = cv2.arcLength(c, True)
    approx = cv2.approxPolyDP(c, 0.02 * perimeter , True)
    if len(approx) == 3:
        shape = "triangle"
    elif len(approx) == 4:
        (x, y, w, h) = cv2.boundingRect(approx)
        ar = w / float(h)
        if (ar >= 0.999 and ar <= 1.001):
            shape = "square" 
        else:
            shape = "rectangle"
    elif len(approx) == 5:
        shape = "pentagon"
    elif len(approx) == 6:
        shape = "hexagon"
    else:
        shape = "circle"
    return shape


# In[ ]:


d=detect(contours[1])
print(d)


# Below 'Dimensioning()' function stores the image temprory amd gives Image imformation like its 'Shape' and 'Ratio'

# In[ ]:


def Dimensioning(userId,view,image):
    os.makedirs('static/temp', exist_ok=True)
    img = cv2.imread(image,cv2.IMREAD_UNCHANGED)
    ratio = 0
    shape = "unidentified"
    try:
        w,h,_=img.shape
        drawSize = int(h/300)
        imgrey = cv2.cvtColor(img.copy(),cv2.COLOR_BGR2GRAY)
        ret, thresh = cv2.threshold(imgrey,127,255,cv2.THRESH_BINARY_INV)
        contours, hierarchy = cv2.findContours(thresh,cv2.RETR_CCOMP,cv2.CHAIN_APPROX_SIMPLE)
        hierarchy,contours = zip(*sorted(zip(hierarchy[0],contours),key = lambda x: cv2.contourArea(x[1]),reverse=True))
        for i,c in enumerate(contours):
            if(hierarchy[i][3] != -1  or (hierarchy[i][3] == -1 and hierarchy[i][2] == -1) ):
                M = cv2.moments(c)
                if(M["m00"] !=0):
                    cX = int(M["m10"] / M["m00"])
                    cY = int(M["m01"] / M["m00"])
                    rect = cv2.minAreaRect(c)
                    box = cv2.boxPoints(rect)
                    box = np.int0(box)
                    x,y,w,h = cv2.boundingRect(c)
                    shape, cylinder_type = detect(c)              
                    if(shape == "unidentified"):
                        continue
                    if(shape=="triangle" or shape=="pentagon" or shape=="hexagon"):
                        img = cv2.drawContours(img, [box], 0, (0,0, 255), drawSize)
                    if(shape=="circle"):
                        img = cv2.rectangle(img,(x,y),(x+w,y+h),(0, 0, 255),drawSize)
                        cv2.line(img, (x,y), (x+w,y), (0,255, 0), 2)
                        ratio = 1/w
                    else:
                        cv2.line(img, tuple(box[0]), tuple(box[1]), (0,255, 0), 2)
                        ratio = 1.0/rect[1][1]
                    break
        folder = 'static/temp/' + userId
        os.makedirs(folder, exist_ok=True)
        try:
            os.remove(folder + '/' + view + '.jpg')
        except: pass
        path_file = (folder + '/' + view + '.jpg')
        small = cv2.resize(img, (0,0), fx=0.5, fy=0.5) 
        cv2.imwrite(path_file,small)
        ratio = str(ratio)
        data={'image': path_file,'shape': shape,'ratio': ratio}
        display(Image(filename = path_file))
        return data
    except:
        print(view + "image not found")


# In[ ]:


def Convert(userId, front_image, side_image, top_image, fratio, sratio, tratio): 
    start = time.time() 
    os.makedirs('static/'+userId, exist_ok=True)
    img_front = cv2.imread(front_image,cv2.IMREAD_UNCHANGED)
    img_side = cv2.imread(side_image,cv2.IMREAD_UNCHANGED)
    img_top = cv2.imread(top_image,cv2.IMREAD_UNCHANGED)
    fratio = float(fratio)
    sratio = float(sratio)
    tratio = float(tratio)
    filePath = "static/temp/"+userId
    try:
        shutil.rmtree(filePath)
    except OSError as e:
        print ("Error: %s - %s." % (e.filename, e.strerror))
    primitive = []
    object_front = valid_contours(img_front,"front",fratio)
    re_arrange(object_front,"front")
    object_side = valid_contours(img_side,"side",sratio)
    re_arrange(object_side,"side")
    object_top = valid_contours(img_top,"top",tratio)
    re_arrange(object_top,"top")
    minApprox = 0.05
    primitive = combining(object_front,object_side,object_top,minApprox)
    final = []
    for set in primitive:
        for shape in set:
            final.append(shape[0])
    try:
        os.remove('static/' + userId + "/" + userId + '.scad')
    except: pass
    path_file = ('static/' + userId + "/" + userId + '.scad')
    if(len(final) == 0):
        path_file = 'static/error.txt'
        f = open(path_file, "w")
        f.write("Cannot determine the 3d geometry, check your files again!")
        f.close()
    createtree(final,path_file)
    end = time.time() 
    print("Total time taken to convert:",end-start)
    return path_file


# In[ ]:


userId="1"
front_image = "TestBench/"+userId+"/front.jpg"
side_image = "TestBench/"+userId+"/side.jpg"
top_image = "TestBench/"+userId+"/top.jpg"
front_image


# In[ ]:


d1 = Dimensioning(userId,"front",front_image)
d1


# In[ ]:


d2 = Dimensioning(userId,"side",side_image)
d2


# In[ ]:


d3 = Dimensioning(userId,"top",top_image)
d3


# In[ ]:


fratio = float(d1["ratio"]) * 2 
sratio = float(d2["ratio"]) * 2
tratio = float(d3["ratio"]) * 2


# Convert Function joins the multiple 2D images with their respective Ratio to insert them in a tree to form a 3D model from it .
# If the given shape consist of more than one shape (complex object in Image object) Than we will sub divide it to individual shapes and store them in tree as Parent and Child . 

# In[ ]:


Convert(userId,front_image, side_image, top_image,fratio, sratio, tratio)


# Point Cloud Representation of the generated Scad File 
# 

# In[ ]:


print("please wait generating 3D view...") 
start = time.time() 
scad_filename = "static/" + userId + '/' + userId + ".1.scad"
stl_filename = "static/" + userId + '/' + userId + ".stl"
pcd_filename = "static/" + userId + '/' + userId + ".pcd"
if(os.path.isfile(scad_filename)):
    run("openscad.exe -o " + stl_filename + " " + scad_filename)
    if(os.path.isfile(stl_filename)):
        print("stl file generated at " + stl_filename)
        mesh = o3d.io.read_triangle_mesh(stl_filename)
        pointcloud = mesh.sample_points_poisson_disk(100000)
        o3d.io.write_point_cloud(pcd_filename, pointcloud)
        model = PyntCloud.from_file(pcd_filename)
        print("pcd file generated at " + pcd_filename)
        model.plot() 
        end = time.time() 
        print("Total time taken to convert from stl to point cloud:",end-start)
    else:
        print("file not found")
else:
    print("scad file not found")