Added EDA file

app.py

@@ -0,0 +1,78 @@
+import streamlit as st #To create web interface for YOLO
+import cv2 #for img and video processing
+import numpy as np #to convert img bytes into numpy arrays
+import tempfile #to handle uploaded video files temporarily
+import time  #to control frame rate when processing videos
+from ultralytics import YOLO
+
+#Loading the pretrained model
+model=YOLO('best.pt')
+
+#Sidebar configuration
+st.sidebar.title("Settings")
+confidence_threshold=st.sidebar.slider("Confidence Threshold",0.0,1.0,0.5,0.05)
+mode=st.sidebar.radio("Select Mode",["Image","Video","Webcam"])
+
+#main configuration
+st.title("🔍 YOLO Object Recognition Sytem")
+if mode=="Image":
+    st.header("🖼️ Image Detection")
+    uploaded_image=st.file_uploader("Upload an image",type=["jpg","jpeg","png"])
+    if uploaded_image is not None:
+        file_bytes=np.asarray(bytearray(uploaded_image.read()),dtype=np.uint8) #returns the numpy array of bytes->pixels
+        image=cv2.imdecode(file_bytes,cv2.IMREAD_COLOR) #numpy array of pixels to image
+        st.image(image,channels="BGR",caption="Original Image") #Display original image
+        results=model.predict(source=image,conf=confidence_threshold,show=False)#making predictions
+        annotated_image=results[0].plot() #Returns the image with detected objects drawn on it.
+        st.image(annotated_image,channels='BGR',caption='Detection Result')
+elif mode=="Video":
+    st.header("🎥 Video Detection")
+    uploaded_video = st.file_uploader("Upload a Video", type=["mp4", "mov", "avi"])
+    if uploaded_video is not None:
+        tfile=tempfile.NamedTemporaryFile(delete=False)
+        tfile.write(uploaded_video.read())
+        cap=cv2.VideoCapture(tfile.name)
+        stframe=st.empty()
+        st.write("Processing Video...⏳")
+        while cap.isOpened():
+            ret,frame=cap.read()
+            if not ret:
+                break
+            #yolo detection
+            results=model.predict(source=frame,conf=confidence_threshold,show=False)
+            annotated_frame=results[0].plot()
+            #display vide frames
+            stframe.image(annotated_frame,channels="BGR")
+            time.sleep(0.03)#Control frame rate
+        cap.release()
+        st.success("Video Processing Complete ✅")
+elif mode=="Webcam":
+    st.header("📷 Live Webcam Detection")
+    # Webcam session control
+    if "run_webcam" not in st.session_state:
+        st.session_state.run_webcam = False
+        #Uses st.session_state to store whether the webcam is running or not.
+    col1,col2=st.columns(2)
+    with col1:
+        start_button=st.button("▶ Start Webcam")
+    with col2:
+        stop_button = st.button("⏹ Stop Webcam")
+    if start_button:
+        st.session_state.run_webcam = True
+    if stop_button:
+        st.session_state.run_webcam = False
+    webcam_placeholder = st.empty()
+    if st.session_state.run_webcam:
+        cap = cv2.VideoCapture(0)  # Open default webcam
+        st.write("Webcam is running... 🎥")
+        while st.session_state.run_webcam and cap.isOpened():
+              ret, frame = cap.read()
+              if not ret:
+                st.error("❌ Failed to grab frame")
+                break
+              results = model.predict(source=frame, conf=confidence_threshold, show=False)
+              annotated_frame = results[0].plot()
+              webcam_placeholder.image(annotated_frame, channels="BGR")
+              time.sleep(0.03)
+        cap.release()
+        st.warning("Webcam Stopped 🛑")

final.py

@@ -0,0 +1,83 @@
+import streamlit as st
+import cv2
+import numpy as np
+import tempfile
+import time
+from ultralytics import YOLO
+
+# Sidebar: Select Model Type
+task = st.sidebar.radio("Choose Task", ["Object Detection", "Instance Segmentation"])
+
+# Load the appropriate model
+if task == "Object Detection":
+    model = YOLO('best.pt')  # Detection model
+    st.title("🔍 YOLO Object Detection System")
+else:
+    model = YOLO('yolo11n-seg.pt')  # Segmentation model
+    st.title("🔍 YOLO Object Segmentation System")
+
+# Sidebar Configuration
+st.sidebar.title("Settings")
+confidence_threshold = st.sidebar.slider("Confidence Threshold", 0.0, 1.0, 0.5, 0.05)
+mode = st.sidebar.radio("Select Mode", ["Image", "Video", "Webcam"])
+
+# Main Logic
+if mode == "Image":
+    st.header("🖼️ Image Processing")
+    uploaded_image = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
+    if uploaded_image is not None:
+        file_bytes = np.asarray(bytearray(uploaded_image.read()), dtype=np.uint8)
+        image = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
+        st.image(image, channels="BGR", caption="Original Image")
+        results = model.predict(source=image, conf=confidence_threshold, show=False)
+        annotated_image = results[0].plot()
+        st.image(annotated_image, channels='BGR', caption='Processed Result')
+
+elif mode == "Video":
+    st.header("🎥 Video Processing")
+    uploaded_video = st.file_uploader("Upload a Video", type=["mp4", "mov", "avi"])
+    if uploaded_video is not None:
+        tfile = tempfile.NamedTemporaryFile(delete=False)
+        tfile.write(uploaded_video.read())
+        cap = cv2.VideoCapture(tfile.name)
+        stframe = st.empty()
+        st.write("Processing Video...⏳")
+        while cap.isOpened():
+            ret, frame = cap.read()
+            if not ret:
+                break
+            results = model.predict(source=frame, conf=confidence_threshold, show=False)
+            annotated_frame = results[0].plot()
+            stframe.image(annotated_frame, channels="BGR")
+            time.sleep(0.03)
+        cap.release()
+        st.success("Video Processing Complete ✅")
+
+elif mode == "Webcam":
+    st.header("📷 Live Webcam Processing")
+    if "run_webcam" not in st.session_state:
+        st.session_state.run_webcam = False
+    col1, col2 = st.columns(2)
+    with col1:
+        start_button = st.button("▶ Start Webcam")
+    with col2:
+        stop_button = st.button("⏹ Stop Webcam")
+    if start_button:
+        st.session_state.run_webcam = True
+    if stop_button:
+        st.session_state.run_webcam = False
+    webcam_placeholder = st.empty()
+    if st.session_state.run_webcam:
+        cap = cv2.VideoCapture(0)
+        st.write("Webcam is running... 🎥")
+        while st.session_state.run_webcam and cap.isOpened():
+            ret, frame = cap.read()
+            if not ret:
+                st.error("❌ Failed to grab frame")
+                break
+            results = model.predict(source=frame, conf=confidence_threshold, show=False)
+            annotated_frame = results[0].plot()
+            webcam_placeholder.image(annotated_frame, channels="BGR")
+            time.sleep(0.03)
+        cap.release()
+        st.warning("Webcam Stopped 🛑")