Update app.py
Browse files
app.py
CHANGED
|
@@ -7,19 +7,21 @@ import yolov9
|
|
| 7 |
HTML_TEMPLATE = """
|
| 8 |
<style>
|
| 9 |
body {
|
| 10 |
-
background: linear-gradient(135deg, #
|
| 11 |
-
font-family: '
|
| 12 |
color: #ecf0f1;
|
|
|
|
| 13 |
}
|
| 14 |
#app-header {
|
| 15 |
text-align: center;
|
| 16 |
-
background: rgba(
|
| 17 |
-
padding:
|
| 18 |
border-radius: 20px;
|
| 19 |
-
box-shadow: 0
|
| 20 |
position: relative;
|
| 21 |
overflow: hidden;
|
| 22 |
-
margin-bottom:
|
|
|
|
| 23 |
}
|
| 24 |
#app-header::before {
|
| 25 |
content: "";
|
|
@@ -28,89 +30,104 @@ HTML_TEMPLATE = """
|
|
| 28 |
left: -50%;
|
| 29 |
width: 200%;
|
| 30 |
height: 200%;
|
| 31 |
-
background: radial-gradient(circle, rgba(
|
| 32 |
-
animation: shimmer
|
| 33 |
}
|
| 34 |
@keyframes shimmer {
|
| 35 |
0% { transform: rotate(0deg); }
|
| 36 |
100% { transform: rotate(360deg); }
|
| 37 |
}
|
| 38 |
#app-header h1 {
|
| 39 |
-
color: #
|
| 40 |
-
font-size:
|
| 41 |
-
margin-bottom:
|
| 42 |
-
text-shadow: 2px 2px 4px rgba(0,0,0,0.
|
| 43 |
}
|
| 44 |
#app-header p {
|
| 45 |
-
font-size: 1.
|
| 46 |
-
color: #
|
| 47 |
}
|
| 48 |
.feature-container {
|
| 49 |
display: flex;
|
| 50 |
justify-content: center;
|
| 51 |
-
gap:
|
| 52 |
-
margin-top:
|
| 53 |
flex-wrap: wrap;
|
| 54 |
}
|
| 55 |
.feature {
|
| 56 |
position: relative;
|
| 57 |
-
transition:
|
| 58 |
border-radius: 15px;
|
| 59 |
overflow: hidden;
|
| 60 |
-
background:
|
| 61 |
-
box-shadow: 0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
}
|
| 63 |
.feature:hover {
|
| 64 |
-
transform: translateY(-
|
| 65 |
-
box-shadow: 0
|
|
|
|
| 66 |
}
|
| 67 |
.feature-icon {
|
| 68 |
-
font-size:
|
| 69 |
-
|
| 70 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
}
|
| 72 |
.feature-description {
|
| 73 |
-
background-color: #34495e;
|
| 74 |
color: #ecf0f1;
|
| 75 |
-
|
| 76 |
-
font-size: 0.9em;
|
| 77 |
text-align: center;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 78 |
}
|
| 79 |
.artifact {
|
| 80 |
position: absolute;
|
| 81 |
-
background: radial-gradient(circle, rgba(
|
| 82 |
border-radius: 50%;
|
| 83 |
opacity: 0.5;
|
|
|
|
| 84 |
}
|
| 85 |
.artifact.large {
|
| 86 |
-
width:
|
| 87 |
-
height:
|
| 88 |
-
top: -
|
| 89 |
-
left: -
|
| 90 |
-
animation: float
|
| 91 |
}
|
| 92 |
.artifact.medium {
|
| 93 |
-
width:
|
| 94 |
-
height:
|
| 95 |
-
bottom: -
|
| 96 |
-
right: -
|
| 97 |
-
animation: float
|
| 98 |
}
|
| 99 |
.artifact.small {
|
| 100 |
-
width:
|
| 101 |
-
height:
|
| 102 |
top: 50%;
|
| 103 |
left: 50%;
|
| 104 |
transform: translate(-50%, -50%);
|
| 105 |
-
animation: pulse
|
| 106 |
}
|
| 107 |
@keyframes float {
|
| 108 |
0%, 100% { transform: translateY(0) rotate(0deg); }
|
| 109 |
-
50% { transform: translateY(-
|
| 110 |
}
|
| 111 |
@keyframes pulse {
|
| 112 |
-
0% { transform: scale(1); opacity: 0.5; }
|
| 113 |
-
100% { transform: scale(1.
|
| 114 |
}
|
| 115 |
</style>
|
| 116 |
<div id="app-header">
|
|
@@ -118,28 +135,29 @@ HTML_TEMPLATE = """
|
|
| 118 |
<div class="artifact medium"></div>
|
| 119 |
<div class="artifact small"></div>
|
| 120 |
<h1>YOLOv9: Manhole Detector</h1>
|
| 121 |
-
<p>
|
| 122 |
<div class="feature-container">
|
| 123 |
<div class="feature">
|
| 124 |
-
<div class="feature-icon"
|
| 125 |
-
<div class="feature-description">High
|
| 126 |
</div>
|
| 127 |
<div class="feature">
|
| 128 |
<div class="feature-icon">⚡</div>
|
| 129 |
-
<div class="feature-description">Fast Processing</div>
|
| 130 |
</div>
|
| 131 |
<div class="feature">
|
| 132 |
<div class="feature-icon">🖼️</div>
|
| 133 |
-
<div class="feature-description">Image
|
| 134 |
</div>
|
| 135 |
<div class="feature">
|
| 136 |
-
<div class="feature-icon"
|
| 137 |
-
<div class="feature-description">
|
| 138 |
</div>
|
| 139 |
</div>
|
| 140 |
</div>
|
| 141 |
"""
|
| 142 |
|
|
|
|
| 143 |
def yolov9_inference(img_path, image_size, conf_threshold, iou_threshold):
|
| 144 |
model = yolov9.load('./best.pt')
|
| 145 |
model.conf = conf_threshold
|
|
|
|
| 7 |
HTML_TEMPLATE = """
|
| 8 |
<style>
|
| 9 |
body {
|
| 10 |
+
background: linear-gradient(135deg, #1a2a6c, #b21f1f, #fdbb2d);
|
| 11 |
+
font-family: 'Roboto', sans-serif;
|
| 12 |
color: #ecf0f1;
|
| 13 |
+
min-height: 100vh;
|
| 14 |
}
|
| 15 |
#app-header {
|
| 16 |
text-align: center;
|
| 17 |
+
background: rgba(26, 42, 108, 0.8);
|
| 18 |
+
padding: 40px;
|
| 19 |
border-radius: 20px;
|
| 20 |
+
box-shadow: 0 15px 30px rgba(0, 0, 0, 0.4);
|
| 21 |
position: relative;
|
| 22 |
overflow: hidden;
|
| 23 |
+
margin-bottom: 40px;
|
| 24 |
+
backdrop-filter: blur(10px);
|
| 25 |
}
|
| 26 |
#app-header::before {
|
| 27 |
content: "";
|
|
|
|
| 30 |
left: -50%;
|
| 31 |
width: 200%;
|
| 32 |
height: 200%;
|
| 33 |
+
background: radial-gradient(circle, rgba(253,187,45,0.2) 0%, rgba(253,187,45,0) 70%);
|
| 34 |
+
animation: shimmer 20s infinite linear;
|
| 35 |
}
|
| 36 |
@keyframes shimmer {
|
| 37 |
0% { transform: rotate(0deg); }
|
| 38 |
100% { transform: rotate(360deg); }
|
| 39 |
}
|
| 40 |
#app-header h1 {
|
| 41 |
+
color: #fdbb2d;
|
| 42 |
+
font-size: 3em;
|
| 43 |
+
margin-bottom: 20px;
|
| 44 |
+
text-shadow: 2px 2px 4px rgba(0,0,0,0.3);
|
| 45 |
}
|
| 46 |
#app-header p {
|
| 47 |
+
font-size: 1.3em;
|
| 48 |
+
color: #ecf0f1;
|
| 49 |
}
|
| 50 |
.feature-container {
|
| 51 |
display: flex;
|
| 52 |
justify-content: center;
|
| 53 |
+
gap: 40px;
|
| 54 |
+
margin-top: 40px;
|
| 55 |
flex-wrap: wrap;
|
| 56 |
}
|
| 57 |
.feature {
|
| 58 |
position: relative;
|
| 59 |
+
transition: all 0.4s ease;
|
| 60 |
border-radius: 15px;
|
| 61 |
overflow: hidden;
|
| 62 |
+
background: rgba(178, 31, 31, 0.7);
|
| 63 |
+
box-shadow: 0 8px 20px rgba(0,0,0,0.3);
|
| 64 |
+
width: 180px;
|
| 65 |
+
height: 180px;
|
| 66 |
+
display: flex;
|
| 67 |
+
flex-direction: column;
|
| 68 |
+
justify-content: center;
|
| 69 |
+
align-items: center;
|
| 70 |
}
|
| 71 |
.feature:hover {
|
| 72 |
+
transform: translateY(-15px) rotate(5deg) scale(1.05);
|
| 73 |
+
box-shadow: 0 20px 40px rgba(0,0,0,0.4);
|
| 74 |
+
background: rgba(253, 187, 45, 0.8);
|
| 75 |
}
|
| 76 |
.feature-icon {
|
| 77 |
+
font-size: 4em;
|
| 78 |
+
color: #ecf0f1;
|
| 79 |
+
margin-bottom: 15px;
|
| 80 |
+
transition: all 0.4s ease;
|
| 81 |
+
}
|
| 82 |
+
.feature:hover .feature-icon {
|
| 83 |
+
transform: scale(1.2);
|
| 84 |
}
|
| 85 |
.feature-description {
|
|
|
|
| 86 |
color: #ecf0f1;
|
| 87 |
+
font-size: 1em;
|
|
|
|
| 88 |
text-align: center;
|
| 89 |
+
padding: 0 10px;
|
| 90 |
+
transition: all 0.4s ease;
|
| 91 |
+
}
|
| 92 |
+
.feature:hover .feature-description {
|
| 93 |
+
font-weight: bold;
|
| 94 |
}
|
| 95 |
.artifact {
|
| 96 |
position: absolute;
|
| 97 |
+
background: radial-gradient(circle, rgba(253,187,45,0.3) 0%, rgba(253,187,45,0) 70%);
|
| 98 |
border-radius: 50%;
|
| 99 |
opacity: 0.5;
|
| 100 |
+
filter: blur(40px);
|
| 101 |
}
|
| 102 |
.artifact.large {
|
| 103 |
+
width: 600px;
|
| 104 |
+
height: 600px;
|
| 105 |
+
top: -200px;
|
| 106 |
+
left: -300px;
|
| 107 |
+
animation: float 30s infinite ease-in-out;
|
| 108 |
}
|
| 109 |
.artifact.medium {
|
| 110 |
+
width: 400px;
|
| 111 |
+
height: 400px;
|
| 112 |
+
bottom: -200px;
|
| 113 |
+
right: -200px;
|
| 114 |
+
animation: float 25s infinite ease-in-out reverse;
|
| 115 |
}
|
| 116 |
.artifact.small {
|
| 117 |
+
width: 200px;
|
| 118 |
+
height: 200px;
|
| 119 |
top: 50%;
|
| 120 |
left: 50%;
|
| 121 |
transform: translate(-50%, -50%);
|
| 122 |
+
animation: pulse 8s infinite alternate;
|
| 123 |
}
|
| 124 |
@keyframes float {
|
| 125 |
0%, 100% { transform: translateY(0) rotate(0deg); }
|
| 126 |
+
50% { transform: translateY(-30px) rotate(15deg); }
|
| 127 |
}
|
| 128 |
@keyframes pulse {
|
| 129 |
+
0% { transform: scale(1) translate(-50%, -50%); opacity: 0.5; }
|
| 130 |
+
100% { transform: scale(1.2) translate(-50%, -50%); opacity: 0.8; }
|
| 131 |
}
|
| 132 |
</style>
|
| 133 |
<div id="app-header">
|
|
|
|
| 135 |
<div class="artifact medium"></div>
|
| 136 |
<div class="artifact small"></div>
|
| 137 |
<h1>YOLOv9: Manhole Detector</h1>
|
| 138 |
+
<p>Unleash the power of AI to detect manholes with precision</p>
|
| 139 |
<div class="feature-container">
|
| 140 |
<div class="feature">
|
| 141 |
+
<div class="feature-icon">🎯</div>
|
| 142 |
+
<div class="feature-description">High Precision Detection</div>
|
| 143 |
</div>
|
| 144 |
<div class="feature">
|
| 145 |
<div class="feature-icon">⚡</div>
|
| 146 |
+
<div class="feature-description">Lightning-Fast Processing</div>
|
| 147 |
</div>
|
| 148 |
<div class="feature">
|
| 149 |
<div class="feature-icon">🖼️</div>
|
| 150 |
+
<div class="feature-description">Dynamic Image Resizing</div>
|
| 151 |
</div>
|
| 152 |
<div class="feature">
|
| 153 |
+
<div class="feature-icon">🔧</div>
|
| 154 |
+
<div class="feature-description">Fine-Tuned Thresholds</div>
|
| 155 |
</div>
|
| 156 |
</div>
|
| 157 |
</div>
|
| 158 |
"""
|
| 159 |
|
| 160 |
+
# The rest of the Python code remains the same
|
| 161 |
def yolov9_inference(img_path, image_size, conf_threshold, iou_threshold):
|
| 162 |
model = yolov9.load('./best.pt')
|
| 163 |
model.conf = conf_threshold
|