utils.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

from ultralytics import YOLO
import streamlit as st
from PIL import Image
import config

@st.cache_resource
def load_model(model_path):
    """
    Loads a YOLO object detection model from the specified model_path.
    Parameters:
        model_path (str): The path to the YOLO model file.
    Returns:
        A YOLO object detection model.
    """
    model = YOLO(model_path)
    return model

# Updated Mapping Template for skin conditions
CLASS_NAMES = {
    0: "Acne",
    1: "Pimples",
    2: "Acne Scars",
    3: "Blackhead",
    4: "Cystic",
    5: "Flat Wart",
    6: "Folliculitis",
    7: "Keloid",
    8: "Milium",
    9: "Papular",
    10: "Purulent",
    11: "Sebo-Crystan-Conglo",
    12: "Whitehead"
}

# Product recommendations mapping
PRODUCT_RECOMMENDATIONS = {
    "Acne": "Salicylic acid cleanser, Non-comedogenic moisturizer",
    "Pimples": "Benzoyl peroxide spot treatment, Oil-free sunscreen",
    "Acne Scars": "Vitamin C serum, Hyaluronic acid",
    "Blackhead": "Charcoal mask, Pore strips",
    "Cystic": "Tea tree oil, Spot patches",
    "Flat Wart": "Over-the-counter salicylic acid",
    "Folliculitis": "Antibacterial wash, Topical cream",
    "Keloid": "Silicone-based scar sheets",
    "Milium": "Retinol cream, Exfoliating cleanser",
    "Papular": "Witch hazel toner, Niacinamide serum",
    "Purulent": "Antiseptic cream, Medicated bandages",
    "Sebo-Crystan-Conglo": "Clay mask, Oil-control moisturizer",
    "Whitehead": "Gentle exfoliating scrub, AHA/BHA toner"
}

# Treatment recommendations mapping
TREATMENT_RECOMMENDATIONS = {
    "Acne": "Regular exfoliation, Avoiding heavy makeup",
    "Pimples": "Gentle cleansing routine, Regular hydration",
    "Acne Scars": "Microneedling, Chemical peels",
    "Blackhead": "Manual extraction by a professional, Laser therapy",
    "Cystic": "Corticosteroid injections, Oral antibiotics",
    "Flat Wart": "Cryotherapy, Electrosurgery",
    "Folliculitis": "Warm compresses, Antibiotic therapy",
    "Keloid": "Corticosteroid injections, Laser treatment",
    "Milium": "Professional extraction, Topical retinoids",
    "Papular": "Blue light therapy, Topical treatments",
    "Purulent": "Incision and drainage, Oral antibiotics",
    "Sebo-Crystan-Conglo": "Isotretinoin therapy, Photodynamic therapy",
    "Whitehead": "Steam and extraction, Preventative skincare routine"
}

def count_objects(boxes):
    counts = {}
    for box in boxes:
        obj_class_index = int(box.cls.item())
        obj_class_name = CLASS_NAMES.get(obj_class_index, f"Class {obj_class_index}")
        counts[obj_class_name] = counts.get(obj_class_name, 0) + 1
    return counts


def display_object_counts(counts, col):
    # Start a container for the cards
    with col.container():
        for obj_class_name, count in counts.items():
            if count > 0:  # Only display if the condition was detected
                # Each card will be in its own column
                card_col1, card_col2 = st.columns([1, 2])
                with card_col1:
                    # Use markdown with HTML to create the card look
                    st.markdown(f"""
                        <div style="background-color: #262730; border-radius: 10px; padding: 20px; margin: 10px 0; box-shadow: 0 2px 4px #484a55;">
                            <h4 style="color: #fafafa; margin: 0;">{obj_class_name}</h4>
                            <h5 style="color: #ff4b4b; margin: 0;">Count: {count}</h5>
                        </div>
                    """, unsafe_allow_html=True)
                with card_col2:
                    st.markdown(f"""
                        <div style="background-color: #262730; border-radius: 10px; padding: 20px; margin: 10px 0; box-shadow: 0 2px 4px #484a55;">
                            <h4 style="color: #ff4b4b; margin: 0;">Recommended Products</h4>
                            <p style="color: #fafafa;">{PRODUCT_RECOMMENDATIONS.get(obj_class_name, 'No products available.')}</p>
                            <h4 style="color: #ff4b4b; margin: 0;">Recommended Treatments</h4>
                            <p style="color: #fafafa;">{TREATMENT_RECOMMENDATIONS.get(obj_class_name, 'No treatments available.')}</p>
                        </div>
                    """, unsafe_allow_html=True)


def infer_uploaded_image(conf, model):
    """
    Execute inference for uploaded image
    :param conf: Confidence of YOLOv8 model
    :param model: An instance of the `YOLOv8` class containing the YOLOv8 model.
    :return: None
    """
    source_img = st.sidebar.file_uploader(
        label="Choose an image...",
        type=("jpg", "jpeg", "png", 'bmp', 'webp')
    )

    col1, col2 = st.columns([1, 2])  # Adjusted for better layout

    if source_img:
        with col1:
            uploaded_image = Image.open(source_img)
            st.image(image=source_img, caption="Uploaded Image", use_column_width=True)

        if st.button("A N A L Y Z E", key="analyze_button"):
            with st.spinner("Running..."):
                res = model.predict(uploaded_image, conf=conf)
                boxes = res[0].boxes
                res_plotted = res[0].plot()[:, :, ::-1]

                with col2:
                    st.image(res_plotted, caption="Analyzed Image", use_column_width=True)
                    object_counts = count_objects(boxes)
                    display_object_counts(object_counts, col2)