app.py

import os
import cv2
import math
import torch
import numpy as np
import gradio as gr
import albumentations
import matplotlib.pyplot as plt
from glob import glob
from PIL import Image
from pytorch_grad_cam import EigenCAM
from models.common import DetectMultiBackend
from albumentations.pytorch import ToTensorV2
from utils.augmentations import letterbox
from utils.plots import Annotator, colors
from pytorch_grad_cam.utils.image import show_cam_on_image, scale_cam_image
from utils.torch_utils import select_device, smart_inference_mode
from utils.general import check_img_size, Profile, non_max_suppression, scale_boxes

weights = "runs/train/best_striped.pt"
data = "data.yaml"
# Load model
device = select_device('cpu')
model = DetectMultiBackend(weights, device=device, dnn=False, data=data, fp16=False)
#target_layers = [model.model.model[-1]]

false_detection_data = glob(os.path.join("false_detection", '*.jpg'))
false_detection_data = [x.replace('\\', '/') for x in false_detection_data]

def resize_image_pil(image, new_width, new_height):

    # Convert to PIL image
    img = Image.fromarray(np.array(image))
    
    # Get original size
    width, height = img.size

    # Calculate scale
    width_scale = new_width / width
    height_scale = new_height / height 
    scale = min(width_scale, height_scale)

    # Resize
    resized = img.resize((int(width*scale), int(height*scale)), Image.NEAREST)
    
    # Crop to exact size
    resized = resized.crop((0, 0, new_width, new_height))

    return resized

def display_false_detection_data(false_detection_data, number_of_samples):
    fig = plt.figure(figsize=(10, 10))

    x_count = 5
    y_count = 1 if number_of_samples <= 5 else math.floor(number_of_samples / x_count)

    for i in range(number_of_samples):
        plt.subplot(y_count, x_count, i + 1)
        img = cv2.imread(false_detection_data[i])
        img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
        plt.imshow(img)
        plt.xticks([])
        plt.yticks([])

    return fig

def inference(input_img, conf_thres, iou_thres, is_false_detection_images=True, num_false_detection_images=10):
    im0 = input_img.copy()
    rgb_img = cv2.resize(im0, (640, 640))
    stride, names, pt = model.stride, model.names, model.pt
    imgsz = check_img_size((640, 640), s=stride)  # check image size

    bs = 1
    # Run inference
    model.warmup(imgsz=(1 if pt or model.triton else bs, 3, *imgsz)) 
    seen, windows, dt = 0, [], (Profile(), Profile(), Profile())

    with dt[0]:
        im = letterbox(input_img, imgsz, stride=stride, auto=True)[0]  # padded resize
        im = im.transpose((2, 0, 1))[::-1]
        im = np.ascontiguousarray(im)
        im = torch.from_numpy(im).to(model.device)
        im = im.half() if model.fp16 else im.float()  # uint8 to fp16/32
        im /= 255  # 0 - 255 to 0.0 - 1.0
        if len(im.shape) == 3:
            im = im[None]  # expand for batch dim
    
    # Inference
    with dt[1]:
        pred = model(im, augment=False, visualize=False)

    # NMS
    with dt[2]:
        pred = non_max_suppression(pred, conf_thres, iou_thres, None, False, max_det=10)

    # Process predictions
    for i, det in enumerate(pred):  # per image
        seen += 1
        annotator = Annotator(im0, line_width=2, example=str(model.names))
        if len(det):
            # Rescale boxes from img_size to im0 size
            det[:, :4] = scale_boxes(im.shape[2:], det[:, :4], im0.shape).round()

            # Write results
            for *xyxy, conf, cls in reversed(det):
                c = int(cls)  # integer class
                label = f'{names[c]} {conf:.2f}'
                annotator.box_label(xyxy, label, color=colors(c, True))
    
    if is_false_detection_images:
        # Plot the misclassified data
        misclassified_images = display_false_detection_data(false_detection_data, number_of_samples=num_false_detection_images)
    else:
        misclassified_images = None

    # cam = EigenCAM(model, target_layers)
    # grayscale_cam = cam(im)[0, :]
    # cam_image = show_cam_on_image(rgb_img, grayscale_cam, use_rgb=True)

    return im0, misclassified_images

title = "YOLOv9 model to detect shirt/tshirt"
description = "A simple Gradio interface to infer on YOLOv9 model and detect tshirt in image"
examples = [["image_1.jpg", 0.25, 0.45, True, 10],
            ["image_2.jpg", 0.25, 0.45, True, 10],
            ["image_3.jpg", 0.25, 0.45, True, 10],
            ["image_4.jpg", 0.25, 0.45, True, 10],
            ["image_5.jpg", 0.25, 0.45, True, 10],
            ["image_6.jpg", 0.25, 0.45, True, 10],
            ["image_7.jpg", 0.25, 0.45, True, 10],
            ["image_8.jpg", 0.25, 0.45, True, 10],
            ["image_9.jpg", 0.25, 0.45, True, 10],
            ["image_10.jpg", 0.25, 0.45, True, 10]]

demo = gr.Interface(inference, 
                    inputs = [gr.Image(width=320, height=320, label="Input Image"), 
                              gr.Slider(0, 1, 0.25, label="Confidence Threshold"),
                              gr.Slider(0, 1, 0.45, label="IoU Thresold"),
                              gr.Checkbox(label="Show False Detection"),
                              gr.Slider(5, 35, value=10, step=5, label="Number of False Detection")],
                    outputs= [gr.Image(width=640, height=640, label="Output"),
                              gr.Plot(label="False Detection")],
                    title=title,
                    description=description,
                    examples=examples)

demo.launch()