Spaces:
Running
Running
Running PPE on person image
Browse files- ppe_compliance.py +59 -8
ppe_compliance.py
CHANGED
|
@@ -3,18 +3,22 @@
|
|
| 3 |
Runs TWO detectors on the full image and reports, per person, which REQUIRED PPE
|
| 4 |
items are MISSING (the "negation" logic):
|
| 5 |
|
| 6 |
-
1.
|
| 7 |
-
shoes/vest/glove), reused from ``ppe_pipeline``.
|
| 8 |
-
2. Person detector — the SAME HF-transformers D-FINE model the "Detect &
|
| 9 |
Classify" tab uses (``ustc-community/dfine-*``), filtered to the COCO/
|
| 10 |
Objects365 "person" class. We reuse that tab's already-loaded model when it
|
| 11 |
matches, so the Space doesn't load a third detector.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
|
| 13 |
For each person, PPE boxes that fall inside (with body-region bands, so a helmet
|
| 14 |
being *held* low isn't counted as *worn*) are associated to that person. A person
|
| 15 |
missing any required item is a VIOLATION. Rendering: green box = OK, red box =
|
| 16 |
-
violation, with a verdict chip ("#0 no helmet, glove").
|
| 17 |
-
``ppe/scripts/compliance.py``
|
|
|
|
| 18 |
"""
|
| 19 |
|
| 20 |
import torch
|
|
@@ -44,6 +48,14 @@ REGION_BANDS = {
|
|
| 44 |
# Person detector defaults to the same model the "Detect & Classify" tab uses.
|
| 45 |
DEFAULT_PERSON_MODEL = "medium-obj2coco"
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
_PERSON = None # (model_key, processor, model, person_ids)
|
| 48 |
|
| 49 |
|
|
@@ -91,6 +103,25 @@ def _contain_frac(inner, outer):
|
|
| 91 |
return inter / area
|
| 92 |
|
| 93 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 94 |
def _color_of(name):
|
| 95 |
return PALETTE[ALL_PPE.index(name) % len(PALETTE)] if name in ALL_PPE else (150, 150, 150)
|
| 96 |
|
|
@@ -116,6 +147,8 @@ def run_ppe_compliance(
|
|
| 116 |
|
| 117 |
im = image.convert("RGB") if isinstance(image, Image.Image) else Image.fromarray(image).convert("RGB")
|
| 118 |
|
|
|
|
|
|
|
| 119 |
# 1) Person detector (reused HF D-FINE) — keep only 'person' above threshold.
|
| 120 |
# Use the model's own device (it may be reused from the Detect & Classify tab,
|
| 121 |
# which could be on GPU); run_dfine moves inputs to that device.
|
|
@@ -123,9 +156,27 @@ def run_ppe_compliance(
|
|
| 123 |
pdev = str(next(mdl.parameters()).device)
|
| 124 |
dets = run_dfine(im, proc, mdl, pdev, person_threshold)
|
| 125 |
persons = [(d["conf"], d["box"]) for d in dets if d["cls"] in person_ids]
|
| 126 |
-
|
| 127 |
-
|
| 128 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 129 |
|
| 130 |
# 3) Assign each PPE detection to the person it is most contained in.
|
| 131 |
people = [{"score": s, "box": box, "present": {}} for s, box in persons]
|
|
|
|
| 3 |
Runs TWO detectors on the full image and reports, per person, which REQUIRED PPE
|
| 4 |
items are MISSING (the "negation" logic):
|
| 5 |
|
| 6 |
+
1. Person detector — the SAME HF-transformers D-FINE model the "Detect &
|
|
|
|
|
|
|
| 7 |
Classify" tab uses (``ustc-community/dfine-*``), filtered to the COCO/
|
| 8 |
Objects365 "person" class. We reuse that tab's already-loaded model when it
|
| 9 |
matches, so the Space doesn't load a third detector.
|
| 10 |
+
2. PPE detector — the fine-tuned D-FINE-M (6 classes: goggles/helmet/mask/
|
| 11 |
+
shoes/vest/glove), reused from ``ppe_pipeline``. Rather than one full-frame
|
| 12 |
+
pass (which shrinks every worker in the 640x640 resize and misses small PPE
|
| 13 |
+
like vests), we run it on EACH person crop and map the boxes back. On this
|
| 14 |
+
image that lifted vest detections from 0 -> 4.
|
| 15 |
|
| 16 |
For each person, PPE boxes that fall inside (with body-region bands, so a helmet
|
| 17 |
being *held* low isn't counted as *worn*) are associated to that person. A person
|
| 18 |
missing any required item is a VIOLATION. Rendering: green box = OK, red box =
|
| 19 |
+
violation, with a verdict chip ("#0 no helmet, glove"). The association /
|
| 20 |
+
negation / rendering logic is ported verbatim from ``ppe/scripts/compliance.py``;
|
| 21 |
+
only the source of PPE boxes changed (per-person crops instead of full frame).
|
| 22 |
"""
|
| 23 |
|
| 24 |
import torch
|
|
|
|
| 48 |
# Person detector defaults to the same model the "Detect & Classify" tab uses.
|
| 49 |
DEFAULT_PERSON_MODEL = "medium-obj2coco"
|
| 50 |
|
| 51 |
+
# Per-person PPE detection: pad each person box before cropping so PPE near the
|
| 52 |
+
# edges isn't clipped; cap the number of crops so crowded frames stay responsive
|
| 53 |
+
# (each crop is one PPE forward pass, CPU on the Space).
|
| 54 |
+
PERSON_CROP_PAD = 0.15
|
| 55 |
+
MAX_PERSON_CROPS = 20
|
| 56 |
+
# Collapse the same physical item seen in two overlapping (padded) crops.
|
| 57 |
+
DEDUP_IOU = 0.55
|
| 58 |
+
|
| 59 |
_PERSON = None # (model_key, processor, model, person_ids)
|
| 60 |
|
| 61 |
|
|
|
|
| 103 |
return inter / area
|
| 104 |
|
| 105 |
|
| 106 |
+
def _box_iou(a, b):
|
| 107 |
+
ix1, iy1 = max(a[0], b[0]), max(a[1], b[1])
|
| 108 |
+
ix2, iy2 = min(a[2], b[2]), min(a[3], b[3])
|
| 109 |
+
inter = max(0.0, ix2 - ix1) * max(0.0, iy2 - iy1)
|
| 110 |
+
union = (a[2] - a[0]) * (a[3] - a[1]) + (b[2] - b[0]) * (b[3] - b[1]) - inter
|
| 111 |
+
return inter / union if union > 0 else 0.0
|
| 112 |
+
|
| 113 |
+
|
| 114 |
+
def _dedup_ppe(dets, iou=DEDUP_IOU):
|
| 115 |
+
"""Drop near-duplicate detections of the SAME class (same item caught in two
|
| 116 |
+
overlapping person crops). Keep the highest-scoring box."""
|
| 117 |
+
kept = []
|
| 118 |
+
for name, s, box in sorted(dets, key=lambda d: -d[1]):
|
| 119 |
+
if any(n == name and _box_iou(box, kb) >= iou for n, _, kb in kept):
|
| 120 |
+
continue
|
| 121 |
+
kept.append((name, s, box))
|
| 122 |
+
return kept
|
| 123 |
+
|
| 124 |
+
|
| 125 |
def _color_of(name):
|
| 126 |
return PALETTE[ALL_PPE.index(name) % len(PALETTE)] if name in ALL_PPE else (150, 150, 150)
|
| 127 |
|
|
|
|
| 147 |
|
| 148 |
im = image.convert("RGB") if isinstance(image, Image.Image) else Image.fromarray(image).convert("RGB")
|
| 149 |
|
| 150 |
+
W, H = im.size
|
| 151 |
+
|
| 152 |
# 1) Person detector (reused HF D-FINE) — keep only 'person' above threshold.
|
| 153 |
# Use the model's own device (it may be reused from the Detect & Classify tab,
|
| 154 |
# which could be on GPU); run_dfine moves inputs to that device.
|
|
|
|
| 156 |
pdev = str(next(mdl.parameters()).device)
|
| 157 |
dets = run_dfine(im, proc, mdl, pdev, person_threshold)
|
| 158 |
persons = [(d["conf"], d["box"]) for d in dets if d["cls"] in person_ids]
|
| 159 |
+
# Keep the most confident persons if the frame is crowded (cap PPE passes).
|
| 160 |
+
persons.sort(key=lambda ps: -ps[0])
|
| 161 |
+
persons = persons[:MAX_PERSON_CROPS]
|
| 162 |
+
|
| 163 |
+
# 2) PPE detector (fine-tuned D-FINE-M), run on EACH person crop so small PPE
|
| 164 |
+
# (vests, distant helmets) isn't lost to the 640x640 full-frame resize. Boxes
|
| 165 |
+
# are mapped back to full-image coords, then de-duplicated across crops.
|
| 166 |
+
ppe = []
|
| 167 |
+
for _conf, pb in persons:
|
| 168 |
+
x1, y1, x2, y2 = pb
|
| 169 |
+
pw, ph = x2 - x1, y2 - y1
|
| 170 |
+
cx1 = max(0, int(x1 - PERSON_CROP_PAD * pw))
|
| 171 |
+
cy1 = max(0, int(y1 - PERSON_CROP_PAD * ph))
|
| 172 |
+
cx2 = min(W, int(x2 + PERSON_CROP_PAD * pw))
|
| 173 |
+
cy2 = min(H, int(y2 + PERSON_CROP_PAD * ph))
|
| 174 |
+
if cx2 <= cx1 or cy2 <= cy1:
|
| 175 |
+
continue
|
| 176 |
+
crop = im.crop((cx1, cy1, cx2, cy2))
|
| 177 |
+
for name, s, b in detect_ppe_boxes(crop, threshold=ppe_threshold):
|
| 178 |
+
ppe.append((name, s, [b[0] + cx1, b[1] + cy1, b[2] + cx1, b[3] + cy1]))
|
| 179 |
+
ppe = _dedup_ppe(ppe)
|
| 180 |
|
| 181 |
# 3) Assign each PPE detection to the person it is most contained in.
|
| 182 |
people = [{"score": s, "box": box, "present": {}} for s, box in persons]
|