def enhance(newpath, factor):
for file in os.listdir():
try:
if os.path.isfile(f"{newpath}\\{file}"):
img = Image.open(file)
newimg = ImageEnhance._Enhance(img).enhance(factor)
newimg.save(f"{newpath}\\{file}")
except Exception as err:
print(err)
def main():
st.set_option('deprecation.showfileUploaderEncoding', False)
activities = [
"Foto Filter", "Feature Detection", "Panorama Generator",
"Document Scanner", "About"
]
option = st.sidebar.selectbox("Select mode :", activities)
#wide = st.sidebar.slider("Size",100,1000)
height = st.sidebar.slider("Height :", 200, 1000, key=1)
width = st.sidebar.slider("Width :", 200, 1000, key=2)
amount = [
"Select Option", "Clockwise 90", "Anticlockwise 90", "Clockwise 180"
]
rotate = st.sidebar.selectbox("Rotate :", amount)
if option == "Foto Filter":
st.title("Foto Editor")
st.write(""" ## Upload Image ## """)
image_file = st.file_uploader("To try the filters upload an Image",
type=['jpg', 'png', 'jpeg'])
activity = [
"Original", "Manual filters", "Sketch", "Smoothen", "B & W",
"GrayScale", "Blurring", "Edges"
]
choices = st.sidebar.radio("Try our automated filters : ", activity)
if image_file is not None:
my_image = Image.open(image_file)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
my_image = my_image.rotate(angle1)
elif rotate == "Anticlockwise 90":
my_image = my_image.rotate(angle2)
elif rotate == "Clockwise 180":
my_image = my_image.rotate(angle3)
if choices == 'Original':
st.text("Original Image")
my_image = np.array(my_image.convert('RGB'))
my_image = cv2.resize(my_image, (width, height))
st.image(my_image)
my_image = Image.fromarray(my_image)
st.markdown(get_image_download_link(my_image),
unsafe_allow_html=True)
elif choices == 'Manual filters':
new_image = my_image
brightness = st.sidebar.slider("Brightness", 1, 5)
enhancer = ImageEnhance.Brightness(new_image)
new_image = enhancer.enhance(brightness)
Contrast = st.sidebar.slider("Contrast", 1, 5)
enhancer = ImageEnhance.Contrast(new_image)
new_image = enhancer.enhance(Contrast)
Sharpness = st.sidebar.slider("Sharpness", 1, 15)
enhancer = ImageEnhance.Sharpness(new_image)
new_image = enhancer.enhance(Sharpness)
Color = st.sidebar.slider("Color Balance", 1, 50)
enhancer = ImageEnhance.Color(new_image)
new_image = enhancer.enhance(Color)
Enhance = st.sidebar.slider("Enhance", 0, 10)
new_image = ImageEnhance._Enhance()
new_image = enhancer.enhance(Enhance)
st.text("Manual filters")
new_image = np.array(new_image.convert('RGB'))
new_image = cv2.resize(new_image, (width, height))
st.image(new_image)
new_image = Image.fromarray(new_image)
st.markdown(get_image_download_link(new_image),
unsafe_allow_html=True)
elif choices == "B & W":
new_image = np.array(my_image.convert('RGB'))
new_image = cv2.cvtColor(new_image, cv2.COLOR_BGR2GRAY)
_, th1 = cv2.threshold(new_image, 127, 255, cv2.THRESH_BINARY)
st.text("B & W")
th1 = cv2.resize(th1, (width, height))
st.image(th1)
result = Image.fromarray(th1)
st.markdown(get_image_download_link(result),
unsafe_allow_html=True)
elif choices == "Sketch":
new_image = np.array(my_image.convert('RGB'))
new_image = cv2.cvtColor(new_image, cv2.COLOR_BGR2GRAY)
th2 = cv2.adaptiveThreshold(new_image, 255,
cv2.ADAPTIVE_THRESH_MEAN_C,
cv2.THRESH_BINARY, 7, 3)
st.text("Sketch")
th2 = cv2.resize(th2, (width, height))
st.image(th2)
result = Image.fromarray(th2)
st.markdown(get_image_download_link(result),
unsafe_allow_html=True)
elif choices == "Smoothen":
new_image = np.array(my_image.convert('RGB'))
kernel = np.ones((5, 5), np.float32) / 25
smooth = cv2.filter2D(new_image, -1, kernel)
st.text("Smoothened Image")
smooth = cv2.resize(smooth, (width, height))
st.image(smooth)
result = Image.fromarray(smooth)
st.markdown(get_image_download_link(result),
unsafe_allow_html=True)
elif choices == 'GrayScale':
new_image = np.array(my_image.convert('RGB'))
new_image = cv2.cvtColor(new_image, cv2.COLOR_BGR2GRAY)
st.text("Gray-scale Image")
new_image = cv2.resize(new_image, (width, height))
st.image(new_image)
result = Image.fromarray(new_image)
st.markdown(get_image_download_link(result),
unsafe_allow_html=True)
elif choices == 'Blurring':
new_image = np.array(my_image.convert('RGB'))
new_image = cv2.GaussianBlur(new_image, (7, 7), 0)
st.text("Gaussian Blur")
new_image = cv2.resize(new_image, (width, height))
st.image(new_image)
result = Image.fromarray(new_image)
st.markdown(get_image_download_link(result),
unsafe_allow_html=True)
elif choices == 'Edges':
new_image = np.array(my_image.convert('RGB'))
new_image = cv2.cvtColor(new_image, cv2.COLOR_BGR2GRAY)
new_image = cv2.Canny(new_image, 100, 100)
st.text("Edge Detect")
new_image = cv2.resize(new_image, (width, height))
st.image(new_image)
result = Image.fromarray(new_image)
st.markdown(get_image_download_link(result),
unsafe_allow_html=True)
elif option == "About":
st.title("Photo Editing Application")
st.subheader("Built using OpenCV and Streamlit")
st.write(''' *** Designed by Divyang Mishra***''')
st.write(
''' *** Linkedin : *** https://www.linkedin.com/in/divyang-mishra-67a7651a0/'''
)
st.write(''' *** Github : *** https://github.com/Divyang0601''')
elif option == "Panorama Generator":
st.title("Create a Panorama using multiple images")
st.subheader("Image Stitching")
options = ["2", "3"]
total = st.sidebar.selectbox("Number of images to be upload", options)
if total == "2":
image_file1 = st.file_uploader("Upload Image 1",
type=['jpg', 'png', 'jpeg'],
key=1)
image_file2 = st.file_uploader("Upload Image 2",
type=['jpg', 'png', 'jpeg'],
key=2)
if (image_file1 and image_file2) is not None:
my_image1 = Image.open(image_file1)
my_image2 = Image.open(image_file2)
new_image1 = np.array(my_image1.convert('RGB'))
new_image2 = np.array(my_image2.convert('RGB'))
new_image1 = cv2.resize(new_image1, (0, 0), None, 0.2, 0.2)
new_image2 = cv2.resize(new_image2, (0, 0), None, 0.2, 0.2)
images = []
images.append(new_image1)
images.append(new_image2)
stitcher = cv2.Stitcher.create()
(status, result) = stitcher.stitch(images)
result = cv2.resize(result, (width, height))
if status == cv2.STITCHER_OK:
st.text("Panorama Generated")
st.image(result)
else:
st.text("Panorama Generation Failed")
result = Image.fromarray(result)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
result = result.rotate(angle1)
elif rotate == "Anticlockwise 90":
result = result.rotate(angle2)
elif rotate == "Clockwise 180":
result = result.rotate(angle3)
st.markdown(get_image_download_link(result),
unsafe_allow_html=True)
elif total == "3":
image_file1 = st.file_uploader("Upload Image 1",
type=['jpg', 'png', 'jpeg'],
key=1)
image_file2 = st.file_uploader("Upload Image 2",
type=['jpg', 'png', 'jpeg'],
key=2)
image_file3 = st.file_uploader("Upload Image 3",
type=['jpg', 'png', 'jpeg'],
key=3)
if (image_file1 and image_file2 and image_file3) is not None:
my_image1 = Image.open(image_file1)
my_image2 = Image.open(image_file2)
my_image3 = Image.open(image_file3)
new_image1 = np.array(my_image1.convert('RGB'))
new_image2 = np.array(my_image2.convert('RGB'))
new_image3 = np.array(my_image3.convert('RGB'))
new_image1 = cv2.resize(new_image1, (0, 0), None, 0.2, 0.2)
new_image2 = cv2.resize(new_image2, (0, 0), None, 0.2, 0.2)
new_image3 = cv2.resize(new_image3, (0, 0), None, 0.2, 0.2)
images = []
images.append(new_image1)
images.append(new_image2)
images.append(new_image3)
stitcher = cv2.Stitcher.create()
(status, result) = stitcher.stitch(images)
result = cv2.resize(result, (width, height))
if status == cv2.STITCHER_OK:
st.text("Panorama Generated")
st.image(result)
else:
st.text("Panorama Generation Failed")
result = Image.fromarray(result)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
result = result.rotate(angle1)
elif rotate == "Anticlockwise 90":
result = result.rotate(angle2)
elif rotate == "Clockwise 180":
result = result.rotate(angle3)
st.markdown(get_image_download_link(result),
unsafe_allow_html=True)
elif option == "Document Scanner":
st.title("Scan your Document here :")
st.subheader("Upload your Image")
image_file = st.file_uploader("Upload Image",
type=['jpg', 'png', 'jpeg'])
activity = ["Original", "Scanned"]
choices = st.sidebar.radio("Select choice : ", activity)
brightness = st.sidebar.slider("Brightness", 1, 5)
Contrast = st.sidebar.slider("Contrast", 1, 5)
Sharpness = st.sidebar.slider("Sharpness", 1, 15)
if image_file is not None:
new_image = Image.open(image_file)
new_image1 = np.array(new_image.convert('RGB'))
ratio = new_image1.shape[0] / 500.0
orig = new_image1.copy()
new_image1 = imutils.resize(new_image1, height=500)
gray = cv2.cvtColor(new_image1, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (5, 5), 0)
edged = cv2.Canny(gray, 75, 200)
cnts = cv2.findContours(edged.copy(), cv2.RETR_LIST,
cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
cnts = sorted(cnts, key=cv2.contourArea, reverse=True)[:5]
for c in cnts:
peri = cv2.arcLength(c, True)
approx = cv2.approxPolyDP(c, 0.02 * peri, True)
if len(approx) == 4:
screenCnt = approx
break
x = approx[0][0][0]
y = approx[0][0][1]
x1 = approx[1][0][0]
y1 = approx[1][0][1]
x2 = approx[2][0][0]
y2 = approx[2][0][1]
x3 = approx[3][0][0]
y3 = approx[3][0][1]
cv2.drawContours(new_image1, [approx], -1, (0, 255, 0), 2)
pts1 = np.float32([[x, y], [x1, y1], [x2, y2], [x3, y3]])
pts2 = np.float32([[width, 0], [0, 0], [0, height],
[width, height]])
matrix = cv2.getPerspectiveTransform(pts1, pts2)
imgOutput = cv2.warpPerspective(new_image1, matrix,
(width, height))
imgOutput = cv2.resize(imgOutput, (width, height))
if choices == "Original":
st.title("Original Image")
enhancer = ImageEnhance.Brightness(new_image)
new_image = enhancer.enhance(brightness)
enhancer = ImageEnhance.Contrast(new_image)
new_image = enhancer.enhance(Contrast)
enhancer = ImageEnhance.Sharpness(new_image)
new_image = enhancer.enhance(Sharpness)
new_image = np.array(new_image.convert('RGB'))
new_image = cv2.resize(new_image, (width, height))
new_image = Image.fromarray(new_image)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
new_image = new_image.rotate(angle1)
elif rotate == "Anticlockwise 90":
new_image = new_image.rotate(angle2)
elif rotate == "Clockwise 180":
new_image = new_image.rotate(angle3)
st.image(new_image)
st.markdown(get_image_download_link(new_image),
unsafe_allow_html=True)
elif choices == "Scanned":
st.title("Scanned Image")
imgOutput = Image.fromarray(imgOutput, 'RGB')
enhancer = ImageEnhance.Brightness(imgOutput)
imgOutput = enhancer.enhance(brightness)
enhancer = ImageEnhance.Contrast(imgOutput)
imgOutput = enhancer.enhance(Contrast)
enhancer = ImageEnhance.Sharpness(imgOutput)
imgOutput = enhancer.enhance(Sharpness)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
imgOutput = imgOutput.rotate(angle1)
elif rotate == "Anticlockwise 90":
imgOutput = imgOutput.rotate(angle2)
elif rotate == "Clockwise 180":
imgOutput = imgOutput.rotate(angle3)
st.image(imgOutput)
st.markdown(get_image_download_link(imgOutput),
unsafe_allow_html=True)
elif option == "Feature Detection":
st.title("Detect Features in your Image")
st.subheader("Upload your image")
image_file = st.file_uploader("Upload Image",
type=['jpg', 'png', 'jpeg'])
activity = ["Original", "Face", "Eyes", "Smiles"]
opt = st.sidebar.radio("Detect Feature: ", activity)
if image_file is not None:
if opt == "Original":
new_image = Image.open(image_file)
st.title("Original Image")
new_image = np.array(new_image.convert('RGB'))
new_image = cv2.resize(new_image, (width, height))
st.image(new_image)
new_image = Image.fromarray(new_image)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
new_image = new_image.rotate(angle1)
elif rotate == "Anticlockwise 90":
new_image = new_image.rotate(angle2)
elif rotate == "Clockwise 180":
new_image = new_image.rotate(angle3)
st.markdown(get_image_download_link(new_image),
unsafe_allow_html=True)
elif opt == "Face":
new_image = Image.open(image_file)
new_image = np.array(new_image.convert('RGB'))
new_image = cv2.resize(new_image, (width, height))
new_image1 = cv2.cvtColor(new_image, cv2.COLOR_RGB2GRAY)
faces = face_cascade.detectMultiScale(new_image1, 1.1, 10)
if faces is ():
print("No faces found")
else:
for (x, y, w, h) in faces:
cv2.rectangle(new_image, (x, y), (x + w, y + h),
(127, 0, 255), 2)
st.title("Face Detected")
new_image = cv2.resize(new_image, (width, height))
st.image(new_image)
st.success("Found {} faces".format(len(faces)))
new_image = Image.fromarray(new_image)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
new_image = new_image.rotate(angle1)
elif rotate == "Anticlockwise 90":
new_image = new_image.rotate(angle2)
elif rotate == "Clockwise 180":
new_image = new_image.rotate(angle3)
st.markdown(get_image_download_link(new_image),
unsafe_allow_html=True)
elif opt == "Eyes":
new_image = Image.open(image_file)
new_image = np.array(new_image.convert('RGB'))
new_image = cv2.resize(new_image, (width, height))
new_image1 = cv2.cvtColor(new_image, cv2.COLOR_RGB2GRAY)
faces = face_cascade.detectMultiScale(new_image1, 1.1, 10)
if faces is ():
print("No eyes found")
else:
for (x, y, w, h) in faces:
roi_gray = new_image1[y:y + h, x:x + h]
roi_color = new_image[y:y + h, x:x + h]
eyes = eye_cascade.detectMultiScale(roi_gray, 1.1, 10)
for (ex, ey, ew, eh) in eyes:
cv2.rectangle(roi_color, (ex, ey),
(ex + ew, ey + eh), (127, 0, 255), 2)
st.title("Eyes Detected")
new_image = cv2.resize(new_image, (width, height))
st.image(new_image)
st.success("Found {} eyes".format(2 * len(faces)))
new_image = Image.fromarray(new_image)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
new_image = new_image.rotate(angle1)
elif rotate == "Anticlockwise 90":
new_image = new_image.rotate(angle2)
elif rotate == "Clockwise 180":
new_image = new_image.rotate(angle3)
st.markdown(get_image_download_link(new_image),
unsafe_allow_html=True)
elif opt == "Smiles":
new_image = Image.open(image_file)
new_image = np.array(new_image.convert('RGB'))
new_image = cv2.resize(new_image, (width, height))
new_image1 = cv2.cvtColor(new_image, cv2.COLOR_RGB2GRAY)
faces = face_cascade.detectMultiScale(new_image1, 1.1, 10)
if faces is ():
print("No Smiles found")
else:
for (x, y, w, h) in faces:
roi_gray = new_image1[y:y + h, x:x + h]
roi_color = new_image[y:y + h, x:x + h]
smiles = smile_cascade.detectMultiScale(
roi_gray, 1.8, 20)
for (sx, sy, sw, sh) in smiles:
cv2.rectangle(roi_color, (sx, sy),
(sx + sw, sy + sh), (127, 0, 255), 2)
st.title("Smile Detected")
new_image = cv2.resize(new_image, (width, height))
st.image(new_image)
st.success("Found {} smiles".format(len(faces)))
new_image = Image.fromarray(new_image)
if rotate == "Selection Option":
pass
elif rotate == "Clockwise 90":
new_image = new_image.rotate(angle1)
elif rotate == "Anticlockwise 90":
new_image = new_image.rotate(angle2)
elif rotate == "Clockwise 180":
new_image = new_image.rotate(angle3)
st.markdown(get_image_download_link(new_image),
unsafe_allow_html=True)