File size: 4,342 Bytes
f284af1
 
 
 
 
 
 
cd5b039
 
f284af1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3e4904b
f284af1
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
from flask import Flask, render_template, request, redirect, url_for, jsonify
import cv2
import numpy as np
from tensorflow.lite.python.interpreter import Interpreter
import os

# Define paths to your model and label files
MODEL_PATH = "detect.tflite"
LABEL_PATH = "labelmap.txt"

# Function to load the TFLite model and labels
def load_model():
    interpreter = Interpreter(model_path=MODEL_PATH)
    interpreter.allocate_tensors()
    input_details = interpreter.get_input_details()
    output_details = interpreter.get_output_details()
    height = input_details[0]['shape'][1]
    width = input_details[0]['shape'][2]

    with open(LABEL_PATH, 'r') as f:
        labels = [line.strip() for line in f.readlines()]

    print(f"Model loaded. Input shape: {input_details[0]['shape']}")
    return interpreter, input_details, output_details, height, width, labels

# Function to preprocess the image for the model
def preprocess_image(image, input_details, height, width):
    image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    image_resized = cv2.resize(image_rgb, (width, height))
    input_data = np.expand_dims(image_resized, axis=0)

    if input_details[0]['dtype'] == np.float32:
        input_data = (np.float32(input_data) - 127.5) / 127.5

    print(f"Image preprocessed: shape {input_data.shape}, dtype {input_data.dtype}")
    return input_data

# Function to perform object detection and draw bounding boxes
def detect_objects(image, interpreter, input_details, output_details, labels):
    input_data = preprocess_image(image, input_details, height, width)
    interpreter.set_tensor(input_details[0]['index'], input_data)
    interpreter.invoke()

    boxes = interpreter.get_tensor(output_details[1]['index'])[0]  # bounding box coordinates
    classes = interpreter.get_tensor(output_details[3]['index'])[0]  # class index
    scores = interpreter.get_tensor(output_details[0]['index'])[0]  # confidence scores

    print(f"Detections: {len(scores)} objects detected")

    for i in range(len(scores)):
        if scores[i] > 0.1:  # confidence threshold
            ymin, xmin, ymax, xmax = boxes[i]
            ymin = int(max(1, ymin * image.shape[0]))
            xmin = int(max(1, xmin * image.shape[1]))
            ymax = int(min(image.shape[0], ymax * image.shape[0]))
            xmax = int(min(image.shape[1], xmax * image.shape[1]))
            cv2.rectangle(image, (xmin, ymin), (xmax, ymax), (0, 255, 0), 2)
            label = f'{labels[int(classes[i])]}: {scores[i] * 100:.2f}%'
            cv2.putText(image, label, (xmin, ymin - 10),
                        cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 255, 255), 2)
            print(f"Object {i}: {label} at [{xmin}, {ymin}, {xmax}, {ymax}]")

    return image

# Initialize the Flask app
app = Flask(__name__, static_folder='static')

# Load the TFLite model and labels
interpreter, input_details, output_details, height, width, labels = load_model()

@app.route('/', methods=['GET', 'POST'])
def upload_and_detect():
    if request.method == 'POST':
        if 'file' not in request.files:
            print("No file part in the request")
            return redirect(request.url)
        file = request.files['file']
        if file.filename == '':
            print("No selected file")
            return redirect(request.url)

        # Read the image file
        image = cv2.imdecode(np.frombuffer(file.read(), np.uint8), cv2.IMREAD_COLOR)
        if image is None:
            print("Failed to read image")
            return redirect(request.url)
        
        print(f"Image uploaded: {file.filename}, shape: {image.shape}")

        # Perform object detection
        processed_image = detect_objects(image, interpreter, input_details, output_details, labels)

        # Ensure the static directory exists
        if not os.path.exists(app.static_folder):
            os.makedirs(app.static_folder)

        # Save processed image
        save_path = os.path.join(app.static_folder, 'detected.jpg')
        cv2.imwrite(save_path, processed_image)
        print(f"Processed image saved at: {save_path}")

        # Send back the path to the processed image
        return jsonify({'image_url': url_for('static', filename='detected.jpg')})

    return render_template('index.html')

if __name__ == '__main__':
    app.run(host='0.0.0.0', port=8000)