import io
import gradio as gr
import matplotlib.pyplot as plt
import requests, validators
import torch
import pathlib
from PIL import Image
from transformers import AutoFeatureExtractor, DetrForObjectDetection, YolosForObjectDetection
import os
# colors for visualization
COLORS = [
[0.000, 0.447, 0.741],
[0.850, 0.325, 0.098],
[0.929, 0.694, 0.125],
[0.494, 0.184, 0.556],
[0.466, 0.674, 0.188],
[0.301, 0.745, 0.933]
]
def make_prediction(img, feature_extractor, model):
inputs = feature_extractor(img, return_tensors="pt")
outputs = model(**inputs)
img_size = torch.tensor([tuple(reversed(img.size))])
processed_outputs = feature_extractor.post_process(outputs, img_size)
return processed_outputs[0]
def fig2img(fig):
buf = io.BytesIO()
fig.savefig(buf)
buf.seek(0)
img = Image.open(buf)
return img
def visualize_prediction(pil_img, output_dict, threshold=0.8, id2label=None):
keep = output_dict["scores"] > threshold
boxes = output_dict["boxes"][keep].tolist()
scores = output_dict["scores"][keep].tolist()
labels = output_dict["labels"][keep].tolist()
if id2label is not None:
labels = [id2label[x] for x in labels]
plt.figure(figsize=(16, 10))
plt.imshow(pil_img)
ax = plt.gca()
colors = COLORS * 100
for score, (xmin, ymin, xmax, ymax), label, color in zip(scores, boxes, labels, colors):
ax.add_patch(plt.Rectangle((xmin, ymin), xmax - xmin, ymax - ymin, fill=False, color=color, linewidth=3))
ax.text(xmin, ymin, f"{label}: {score:0.2f}", fontsize=15, bbox=dict(facecolor="yellow", alpha=0.5))
plt.axis("off")
return fig2img(plt.gcf())
def detect_objects(model_name,url_input,image_input,threshold):
#Extract model and feature extractor
feature_extractor = AutoFeatureExtractor.from_pretrained(model_name)
model = DetrForObjectDetection.from_pretrained(model_name)
image = image_input
#Make prediction
processed_outputs = make_prediction(image, feature_extractor, model)
print(processed_outputs)
#Visualize prediction
viz_img = visualize_prediction(image, processed_outputs, threshold, model.config.id2label)
return viz_img
xxresult=0
def detect_objects2(model_name,url_input,image_input,threshold,type2):
#Extract model and feature extractor
feature_extractor = AutoFeatureExtractor.from_pretrained(model_name)
model = DetrForObjectDetection.from_pretrained(model_name)
image = image_input
#Make prediction
processed_outputs = make_prediction(image, feature_extractor, model)
print(processed_outputs)
#Visualize prediction
viz_img = visualize_prediction(image, processed_outputs, threshold, model.config.id2label)
keep = processed_outputs["scores"] > threshold
det_lab = processed_outputs["labels"][keep].tolist()
det_lab.count(6)
if det_lab.count(6) > 0:
total_text="Trench is Detected \n Image is Not Blurry \n"
else:
total_text="Trench is NOT Detected \n Image is Blurry \n"
print(type2)
print(type(type2))
if det_lab.count(4) > 0:
total_text+="Measuring Tape (Vertical) for measuring Depth is Detected \n"
else:
total_text+="Measuring Tape (Vertical) for measuring Depth is NOT Detected \n"
if det_lab.count(5) > 0:
total_text+="Measuring Tape (Horizontal) for measuring Width is Detected \n"
else:
total_text+="Measuring Tape (Horizontal) for measuring Width is NOT Detected \n"
return total_text
def tott(model_name,url_input,image_input,threshold,type2):
#Extract model and feature extractor
feature_extractor = AutoFeatureExtractor.from_pretrained(model_name)
model = DetrForObjectDetection.from_pretrained(model_name)
image = image_input
#Make prediction
processed_outputs = make_prediction(image, feature_extractor, model)
keep = processed_outputs["scores"] > threshold
det_lab = processed_outputs["labels"][keep].tolist()
xxresult=0
if det_lab.count(6) == 0:
xxresult=1
if det_lab.count(4) == 0:
if type2=="Trench Depth Measurement":
xxresult=1
if det_lab.count(5) == 0:
if type2=="Trench Width Measurement":
xxresult=1
if xxresult==0:
return "The photo is ACCEPTED"
else:
return "The photo is NOT ACCEPTED"
def set_example_image(example: list) -> dict:
return gr.Image.update(value=example[0])
def set_example_url(example: list) -> dict:
return gr.Textbox.update(value=example[0])
title = """<h1 id="title">Object Detection App for POC</h1>"""
description = """
This application can be used as follows:
- Select the model
- Select the type of classification
- Select the photo
- Press Detect
- Press Results
"""
models = ["omarhkh/CutLER-finetuned-11" ,"omarhkh/CutLER-finetuned-12","omarhkh/detr-finetuned-omar8" , "omarhkh/CutLER-finetuned-omar3"]
types_class = ["Trench Depth Measurement", "Trench Width Measurement"]
css = '''
h1#title {
text-align: center;
}
'''
demo = gr.Blocks(css=css)
with demo:
gr.Markdown(title)
gr.Markdown(description)
#gr.Markdown(detect_objects2)
options = gr.Dropdown(value="omarhkh/CutLER-finetuned-11",choices=models,label='Select Object Detection Model',show_label=True)
options2 = gr.Dropdown(value="Trench Depth Measurement",choices=types_class,label='Select Classification Type',show_label=True)
slider_input = gr.Slider(minimum=0.1,maximum=1,value=0.8,label='Prediction Threshold')
with gr.Tabs():
with gr.TabItem('Image Upload'):
with gr.Row():
img_input = gr.Image(type='pil')
img_output_from_upload= gr.Image(shape=(650,650))
with gr.Row():
example_images = gr.Dataset(components=[img_input], samples=[[path.as_posix()] for path in sorted(pathlib.Path('images').rglob('*.jpg'))])
img_but = gr.Button('Detect')
with gr.Blocks():
name = gr.Textbox(label="Final Result")
output = gr.Textbox(label="Reason for the results")
greet_btn = gr.Button("Results")
greet_btn.click(fn=detect_objects2, inputs=[options,img_input,img_input,slider_input,options2], outputs=output, queue=True)
greet_btn.click(fn=tott, inputs=[options,img_input,img_input,slider_input,options2], outputs=name, queue=True)
img_but.click(detect_objects,inputs=[options,img_input,img_input,slider_input],outputs=img_output_from_upload,queue=True)
example_images.click(fn=set_example_image,inputs=[example_images],outputs=[img_input])
demo.launch(enable_queue=True)