def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
img1 = base64_to_pil(request.json)
#cam(img1)
preds = model_predict(img, model)
# Process your result for human
pred_class = decode_predictions(preds) # ImageNet Decode
result = str(pred_class[0][0][1]) # Convert to string
result = result.replace('_', ' ').capitalize()
return jsonify(result=result) #, probability=pred_proba)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
print(request.json)
print('aaa')
img1, img2 = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
preds = model_predict(img1, img2, model)
# Process your result for human
# pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
# result = str(pred_class[0][0][1]) # Convert to string
# result = result.replace('_', ' ').capitalize()
# Serialize the result, you can add additional fields
type(preds)
result = preds
# lists = result.tolist()
# json_str = json.dumps(lists)
# print(json_str)
print(result)
return jsonify(result)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
model_xception = load_model(MODEL_XCEPTION)
img = base64_to_pil(request.json)
preds = model_predict(img, model_xception)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make predictionpy
#preds = model_predict(img, model
# Process your result for human
pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
classes = [
'Abyssinian', 'Bengal', 'Bombay', 'British Shorthair',
'Egyptian Mau', 'Maine Coon', 'Persian', 'Russian Blue', 'Siamese',
'Sphynx'
]
#result = str(classes[0][0][1]) # Convert to string
#result = str(classes[np.argmax(np.array(preds[0]))])
#result = result.replace('_', ' ').capitalize()
result = []
for idx in preds.argsort()[0][::-1][:3]:
arr = classes[idx].split("-")[-1] + ' : ' + "{:.2f}%\n".format(
preds[0][idx] * 100)
#arr = [ '\n{}'.format(x) for x in arr ]
result.append(arr)
a = "Used Model: Xception"
# Serialize the result, you can add additional fields
return jsonify(result=result, prediction=pred_proba, a=a)
return render_template("index.html", a=a)
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
preds = model_predict(img, model)
print("b/m", preds)
# Process your result for human
pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
#pred_class = decode_predictions(preds, top=1) # ImageNet Decode
print("pred_proba", pred_proba)
print(type(preds[0]))
if float(preds[0][1]) >= 0.5:
result = "malignant"
else:
result = "benign"
# Serialize the result, you can add additional fields
return jsonify(result=result, probability=pred_proba)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
preds = model_predict(img, model)
# Process your result for human
if preds == 0:
result = 'No DR'
elif preds == 1:
result = 'Mild'
elif preds == 2:
result = 'Moderate'
elif preds == 3:
result = 'Severe'
else:
result = 'Proliferative DR'
# Serialize the result, you can add additional fields
return jsonify(result=result)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
#preds = model_predict(img, model)
preds = loadModel(img)
# Process your result for human
pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
classes = ['Abyssinian', 'Bengal', 'Bombay', 'British Shorthair', 'Egyptian Mau', 'Maine Coon', 'Persian', 'Russian Blue', 'Siamese', 'Sphynx']
#result = str(classes[0][0][1]) # Convert to string
result = str(classes[np.argmax(np.array(preds[0]))])
result = result.replace('_', ' ').capitalize()
ngetes = []
for idx in preds.argsort()[0][::-1][:5]:
arr = classes[idx].split("\n-")[-1],"\n{:.2f}%\n".format(preds[0][idx]*100)
#arr = [ '\n{}'.format(x) for x in arr ]
ngetes.append(arr)
# Serialize the result, you can add additional fields
return jsonify(result=ngetes, probability=pred_proba)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
preds = model_predict(img, model)
# Process your result for human
pred_proba = "{:.3f}".format(np.max(preds[0], axis=-1)) # Max probability
print("Maximum Probability: ",pred_proba)
# pred_class = decode_predictions(preds, top=1) # ImageNet Decode
# result = str(pred_class[0][0][1]) # Convert to string
# result = result.replace('_', ' ').capitalize()
labels = {0: 'Garbage Image', 1: 'Non-Garbage Image'}
result = labels[np.argmax(preds[0], axis=-1)]
print("Classified:",result)
# Serialize the result, you can add additional fields
return jsonify(result=result, probability=pred_proba)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
preds = model_predict(img, model)
print(preds)
# Process your result for human
pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
print(pred_proba)
threshold = .5
if preds[0] > threshold:
result = 'bacterial'
else:
result = 'viral'
# pred_class = decode_predictions(preds, top=1) # ImageNet Decode
# result = str(pred_class[0][0][1]) # Convert to string
# result = result.replace('_', ' ').capitalize()
# Serialize the result, you can add additional fields
return jsonify(result=result, probability=pred_proba)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
preds = get_embedding(model, img)
f = open('store.pckl', 'rb')
df, df_embs = pickle.load(f)
f.close()
# append extracted features to dataset
df_embs = df_embs.append(pd.Series(preds), ignore_index=True)
# compute cos similarities between objects:
cosine_sim = 1 - pairwise_distances(df_embs, metric='cosine')
# indices = pd.Series(range(len(df)), index=df.index)
# return top-10 similar items for input image. It's index in the dataframe is 5000 (the last row)
#idx_rec, idx_sim = get_recommender(5000, df, cosine_sim, indices, top_n=10)
top_n = 5
sim_scores = list(enumerate(cosine_sim[-1]))
sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)
sim_scores = sim_scores[1:top_n + 1]
idx_rec = [i[0] for i in sim_scores]
idx_sim = [i[1] for i in sim_scores]
out = df[['images.model', 'shortDescription',
'priceInfo.finalPrice']].loc[idx_rec].to_json(orient='index')
# Serialize the result, you can add additional fields
return jsonify(out)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Make prediction
preds = model_predict(img, model)
if preds == 0:
result = "MEL"
elif preds == 1:
result = "NV"
elif preds == 2:
result = "BCC"
elif preds == 3:
result = "AK"
elif preds == 4:
result = "BKL"
elif preds == 5:
result = "DF"
elif preds == 6:
result = "VASC"
elif preds == 7:
result = "SCC"
else:
result = "UNK"
return jsonify(result=result, probability=int(preds))
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json).convert('RGB')
#print(request.files['file'])
# Save the image to ./uploads
#img.save("uploads/image.png")
# Make prediction
preds = model_predict(img, model)
gradcam = gradcam_from_img(img)
# Process your result for human
if (preds[0][0] >= preds[0][1]):
pred_class = "POSITIVE"
else:
pred_class = "NEGATIVE"
pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
# Serialize the result, you can add additional fields
return jsonify(result=pred_class,
probability=pred_proba,
file=img_to_base64(gradcam))
return None
def predict():
try:
img = base64_to_pil(request.json)
except Exception:
return jsonify(error="expected string or bytes-like object"), 422
rgb_image_np = np.array(img.convert('RGB'))
stds = rgb_image_np.std(axis=(0, 1))
means = rgb_image_np.mean(axis=(0, 1))
if 35 < np.average(stds) < 95 and 60 < np.average(means) < 180:
img_result = x_ray.predict(img)
condition_similarity_rate = []
if img_result['condition rate'] == []:
return jsonify(
result='NOT DETECTED',
error='Invalid image'
), 200
else:
for name, prob in img_result['condition rate']:
condition_similarity_rate.append({'y': round(float(prob), 3), 'name': name})
print(condition_similarity_rate)
return jsonify(
condition_similarity_rate=condition_similarity_rate
), 200
else:
file_name = "NOT_DETECTED/%s.jpg" % str(uuid.uuid4())
return jsonify(
result='NOT DETECTED',
error='Invalid image'
), 200
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
preds = model_predict(img, model)
preds = np.argmax(preds, axis=1)
# Process your result for human
pred_class = [
":( Oops, Not a Cat",
":) Yes, It's a Cat!",
] # ImageNet Decode
result = [pred_class[index] for index in preds] # Convert to string
# Serialize the result, you can add additional fields
return jsonify(result=result[0])
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
img.save("uploads\image.jpg")
img_path = os.path.join(os.path.dirname(__file__), 'uploads\image.jpg')
os.path.isfile(img_path)
img = image.load_img(img_path, target_size=(64, 64))
preds = model_predict(img, model)
result = preds[0, 0]
print(result)
if result > 0.5:
return jsonify(result="PNEMONIA")
else:
return jsonify(result="NORMAL")
return None
def predict():
if request.method == 'POST':
# Get the image from post request
image = base64_to_pil(request.json['image'])
height = int(request.json['height'])
if not os.path.exists('../_tmp/in'):
tf.gfile.MakeDirs('../_tmp/in')
image.save('../_tmp/in/work.png')
# res_im,seg=MODEL.run(image)
# seg=cv2.resize(seg.astype(np.uint8),image.size)
# mask_sel=(seg==15).astype(np.float32)
# mask = 255*mask_sel.astype(np.uint8)
# img = np.array(image)
# img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
# res = cv2.bitwise_and(img,img,mask = mask)
# bg_removed = res + (255 - cv2.cvtColor(mask, cv2.COLOR_GRAY2BGR))
# result = main(bg_removed,height,None)
result = demo.run('../_tmp/in', height)
obj = ''
with open('test.obj', 'r') as file:
obj = file.read()
return jsonify(result=result, obj=obj)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
img.save("./uploads/image.png")
# Make prediction
preds = model_predict(img, model1)
switcher = {
0: "Airplane",
1: "auto-mobile",
2: "Bird",
3: "Cat",
4: "Deer",
5: "Dog",
6: "Frog",
7: "Horse",
8: "Ship",
9: "Truck"
}
result = switcher.get(int(preds), "Nothing")
# Serialize the result, you can add additional fields
return jsonify(result=result)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Save the image to ./uploads
# img.save("./uploads/image.png")
# Make prediction
preds = model_predict(img, model)
pred_class = np.argmax(preds, axis=1)
# Process your result for human
pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
#pred_class = decode_predictions(preds, top=1) # ImageNet Decode
if pred_class == 0:
result = 'Not subluxed'
elif pred_class == 1:
result = 'Subluxed'
#result = str(pred_class[0][0][1]) # Convert to string
result = result.replace('_', ' ').capitalize()
# Serialize the result, you can add additional fields
return jsonify(result=result, probability=pred_proba)
return None
def upload_backend(label):
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
img.save("uploads/" + label + "/" + str(secrets.token_hex(32)) +
".png")
return jsonify(result="Thank you!")
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Make prediction
return model_predict(img)
return None
def predict():
if request.method == 'POST':
img1, img2 = base64_to_pil(request.json)
img3 = model_predict(img1, img2)
img3 = cv2.resize(img3, (400, 300), interpolation=cv2.INTER_AREA)
img3 = np_to_base64(img3)
#return jsonify({'image_url': '/output.png'})
return jsonify(img3)
return None
def predict():
if request.method == 'POST':
# print(request.json) # data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/4QAIRXhpZgAA/9sAQwAFAwQEBAM
# # Get the image from post request
# img = base64_to_pil(request.json)
img = base64_to_pil(request.json).convert('L').resize((48, 48))
img = np.asarray(img).reshape(-1, 48, 48, 1).astype('float32') / 255
# print(img.shape)
# # # Save the image to ./uploads
# img.save("./uploads/image.png")
begin = time.time()
p_class = predict_class(img)
predict_time1 = time.time()
p_value = predict_value(img)
predict_time2 = time.time()
print("time:",str(predict_time1-begin),str(predict_time2-begin))
print("预测结果(lable)：", p_class)
print("预测结果：", label_dict[p_class[0]])
print("每个预测值：", p_value)
# 终端显示每个预测结果对应的预测值
# probability_yang = show_predict_result(p_class, p_value)
# prediction = model.predict_classes(img)
# prediction_probability = model.predict(img)
# print(prediction)
# print(prediction_probability)
# # # Make prediction
# preds = model_predict(img, model)
#
# # Process your result for human
# probability = "{:.3f}".format(np.amax(p_class[0])) # Max probability
# pred_class = decode_predictions(preds, top=1) # ImageNet Decode
#
result = str(label_dict[p_class[0]]) # Convert to string
result = result.replace('_', ' ').capitalize()
# probability =str(p_value)
# probability= json.dumps(p_value.item())
# probability=p_value.tolist()
# probability = str(p_value) # Convert to string
# probability = probability.replace('_', ' ').capitalize()
probability = str(",".join(str(float(_*100)) for _ in p_value[0]))
print(result)
print(probability)
# print(result)
# # Serialize the result, you can add additional fields
return jsonify(result=result, probability=probability)
# return jsonify(result=result, probability=probability.yang)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
preds = model_predict(img, test)
result = str(preds)
return jsonify(result=result)
print("end")
return None
def predict():
# 需要从request对象读取表单内容：
if request.method == 'POST':
render_template('index.html')
img = base64_to_pil(request.json)
prediction = model_predict(img, model).numpy().item()
prediction_class = class_map[prediction]
return jsonify(result=prediction_class)
return '<h3>Bad username or password.</h3>'
def predict():
if request.method == 'POST':
img = base64_to_pil(request.json)
hash_value = str(hex(hash(img.tobytes())))
response = inference_handler.predict(img)
storage.temp_store(hash_value, img)
return jsonify(result=response, hash_value=hash_value)
return None
def findEdge():
img = base64_to_pil(request.form['image'])
target = np.array([request.form['x'], request.form['y']], dtype=float)
nearestPoint, distance = find_nearest_white(img, target)
result = {
'x': int(nearestPoint[0]),
'y': int(nearestPoint[1]),
'distance': float(distance)
}
return jsonify(result)
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Make prediction
##preds = model_predict(img, model)
pred_proba = model_predict(img, model)
pred_class = np.argmax(pred_proba)
# Process your result for human
##pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
pred_proba = "{:.3f}".format(np.argmax(pred_proba))
if pred_class < 0.9:
result = 'No_fire'
elif pred_class >= 0.9:
result = 'Fire'
sendEmail()
#Make threshold basis results coming along
result = result.replace('_', ' ').capitalize()
#sendEmail()
# Process your result for human
##pred_proba = "{:.3f}".format(np.amax(preds)) # Max probability
##pred_class = decode_predictions(preds, top=1) # ImageNet Decode
##result = str(pred_class[0][0][1]) # Convert to string
##result = result.replace('_', ' ').capitalize()
# Serialize the result, you can add additional fields
this_image = f"./images/{datetime.now().timestamp()}.png"
################# ACTION: SAVE YOUR MASKED IMAGE HERE ##################
#Watever function you have change the `img.save`
img.save(this_image)
################# ACTION END ##################
image_file = open(this_image, "rb")
image = base64.b64encode(image_file.read())
image_string = str(
image
)[2:
-1] #converting it into string and removing the 'b' and the quotes from the start and end
return jsonify(result=result,
probability=pred_proba,
image=image_string)
return None
def predict():
if request.method == 'POST':
img = base64_to_pil(request.json)
img.save("uploads/" + filename + ".png")
preds = model_predict(img, model)
pred_proba = "{:.3f}".format(np.amax(preds))
pred_class = decode_predictions(preds, top=1)
result = str(pred_class[0][0][1])
result = result.replace('_', ' ').capitalize()
searchterm = result
directory = "./images/"
url = "https://www.google.co.in/search?q=" + searchterm + "&source=lnms&tbm=isch"
browser = webdriver.Chrome('C:\WebDrivers\chromedriver.exe')
browser.get(url)
extensions = {"jpg", "jpeg", "png", "gif"}
if not os.path.exists(directory):
os.mkdir(directory)
for _ in range(500):
browser.execute_script("window.scrollBy(0,10000)")
html = browser.page_source.split('["')
imges = []
for i in html:
if i.startswith('http') and i.split('"')[0].split(
'.')[-1] in extensions:
imges.append(i.split('"')[0])
print(imges)
def save_image(img, directory):
for img in imges:
img_url = img
img_type = img.split('.')[-1]
try:
path = os.path.join(
directory,
searchterm + "_" + str(uuid.uuid4()) + "." + 'jpg')
urllib.request.urlretrieve(img, path)
except Exception as e:
print(e)
save_image(imges, directory)
browser.close()
fantasy_zip = zipfile.ZipFile('C:\\Users\\User\\Desktop\\FYP\\images.zip',
'w')
for folder, subfolders, files in os.walk(
'C:\\Users\\User\\Desktop\\FYP\\images'):
for file in files:
if file.endswith('.jpg'):
fantasy_zip.write(os.path.join(folder, file),
file,
compress_type=zipfile.ZIP_DEFLATED)
fantasy_zip.close()
return jsonify(result=result, probability=pred_proba)
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json)
# Make prediction
preds = model_predict(img, model)
labels = {'glioma': 0, 'meningioma': 1, "no_tumor":2,'pituitary': 3}
prediction = list(labels.keys())[list(labels.values()).index(list(preds)[0])]
result = str(prediction) # Convert to string
result = result.replace('_', ' ').capitalize()
return jsonify(result=result,preds = str(list(preds)))
def predict():
if request.method == 'POST':
img = base64_to_pil(request.json)
print(img.filename)
path = 'image.jpg'
img.save(path)
result = neural.predict_breed(path)
# Serialize the result
return jsonify(result=result)
return None
def predict():
if request.method == 'POST':
# Get the image from post request
img = base64_to_pil(request.json).convert('RGB')
img.save("./uploads/image.jpg")
# Make prediction
#----------IDS PREDICTIONS -----------------
IDs = IDs_Predictions(img)
#-------------------------------------------
return jsonify(IDS=(IDs))
#return jsonify(result=(max_predicted_class, probability=pred_proba))
return None
