Yuan1z0825/nature-skills / skills/nature-image2ppt
nature-image2ppt
FlaggedReconstruct slide images, screenshots, scanned PDFs, or image-only PPTX files as object-level editable PowerPoint. Use for 图片转可编辑PPT、截图还原PPT and diagram reconstruction; not authoring a new deck from research notes.
cli/image2ppt/runtime/split_alpha_components.py
View on GitHub#!/usr/bin/env python3
import argparse
import json
from pathlib import Path
from PIL import Image, ImageDraw, ImageFilter
def parse_names(value, option="--names"):
if not value:
return []
names = [item.strip() for item in value.split(",") if item.strip()]
for name in names:
path = Path(name)
if (
"\x00" in name
or "/" in name
or "\\" in name
or name in {".", ".."}
or path.is_absolute()
or path.drive
or path.name != name
):
raise SystemExit(f"{option} entries must be safe filename components: {name}")
return names
NEIGHBORS = {
4: ((1, 0), (-1, 0), (0, 1), (0, -1)),
8: ((1, 0), (-1, 0), (0, 1), (0, -1), (1, 1), (1, -1), (-1, 1), (-1, -1)),
}
def foreground_mask(alpha, threshold, close_radius):
mask = alpha.point(lambda p: 255 if p > threshold else 0, mode="L")
if close_radius <= 0:
return mask
size = close_radius * 2 + 1
return mask.filter(ImageFilter.MaxFilter(size)).filter(ImageFilter.MinFilter(size))
def component_boxes(mask, min_area, connectivity):
width, height = mask.size
pixels = mask.load()
seen = bytearray(width * height)
components = []
neighbors = NEIGHBORS[connectivity]
for y in range(height):
for x in range(width):
idx = y * width + x
if seen[idx] or pixels[x, y] == 0:
continue
stack = [(x, y)]
seen[idx] = 1
min_x = max_x = x
min_y = max_y = y
area = 0
while stack:
cx, cy = stack.pop()
area += 1
min_x = min(min_x, cx)
max_x = max(max_x, cx)
min_y = min(min_y, cy)
max_y = max(max_y, cy)
for dx, dy in neighbors:
nx = cx + dx
ny = cy + dy
if 0 <= nx < width and 0 <= ny < height:
nidx = ny * width + nx
if not seen[nidx] and pixels[nx, ny] != 0:
seen[nidx] = 1
stack.append((nx, ny))
if area >= min_area:
components.append({"area": area, "box": [min_x, min_y, max_x + 1, max_y + 1]})
return components
def gap_between(a, b):
ax1, ay1, ax2, ay2 = a
bx1, by1, bx2, by2 = b
dx = max(0, max(ax1, bx1) - min(ax2, bx2))
dy = max(0, max(ay1, by1) - min(ay2, by2))
return max(dx, dy)
def union_box(a, b):
return [
min(a[0], b[0]),
min(a[1], b[1]),
max(a[2], b[2]),
max(a[3], b[3]),
]
def box_area(box):
return max(1, box[2] - box[0]) * max(1, box[3] - box[1])
def should_merge(a, b, max_gap, max_union_growth):
if gap_between(a["box"], b["box"]) > max_gap:
return False
union = union_box(a["box"], b["box"])
separate_area = box_area(a["box"]) + box_area(b["box"])
return box_area(union) / max(1, separate_area) <= max_union_growth
def merge_fragments(components, max_gap, max_union_growth):
if max_gap <= 0:
return components
merged = [dict(component, merged_count=1) for component in components]
changed = True
while changed:
changed = False
for i in range(len(merged)):
for j in range(i + 1, len(merged)):
if not should_merge(merged[i], merged[j], max_gap, max_union_growth):
continue
merged[i] = {
"area": merged[i]["area"] + merged[j]["area"],
"box": union_box(merged[i]["box"], merged[j]["box"]),
"merged_count": merged[i]["merged_count"] + merged[j]["merged_count"],
}
del merged[j]
changed = True
break
if changed:
break
return merged
def sort_components(components, mode):
if mode == "area":
return sorted(components, key=lambda item: item["area"], reverse=True)
if mode == "y":
return sorted(components, key=lambda item: (item["box"][1], item["box"][0]))
return sorted(components, key=lambda item: (item["box"][0], item["box"][1]))
def extract_component_asset(image, box, pad, square):
width, height = image.size
left, top, right, bottom = box
left = max(0, left - pad)
top = max(0, top - pad)
right = min(width, right + pad)
bottom = min(height, bottom + pad)
component_image = image.crop((left, top, right, bottom))
if not square:
return component_image, [left, top, right, bottom]
side = max(component_image.size)
canvas = Image.new("RGBA", (side, side), (0, 0, 0, 0))
canvas.alpha_composite(
component_image,
((side - component_image.size[0]) // 2, (side - component_image.size[1]) // 2),
)
return canvas, [left, top, right, bottom]
def write_contact_sheet(items, out_path):
if not items:
return
thumb = 180
label_h = 28
cols = min(5, len(items))
rows = (len(items) + cols - 1) // cols
sheet = Image.new("RGB", (cols * thumb, rows * (thumb + label_h)), "white")
draw = ImageDraw.Draw(sheet)
for index, item in enumerate(items):
row, col = divmod(index, cols)
preview = item["image"].copy()
preview.thumbnail((thumb - 16, thumb - 16))
x = col * thumb + (thumb - preview.width) // 2
y = row * (thumb + label_h) + (thumb - preview.height) // 2
checker = Image.new("RGB", preview.size, "#f4f4f4")
checker.paste(preview, mask=preview.getchannel("A"))
sheet.paste(checker, (x, y))
draw.text((col * thumb + 8, row * (thumb + label_h) + thumb), item["name"], fill="black")
sheet.save(out_path)
def main():
parser = argparse.ArgumentParser(description="Split a transparent asset sheet into component PNG files.")
parser.add_argument("--input", required=True, help="RGBA asset sheet after chroma-key removal")
parser.add_argument("--out-dir", required=True, help="Directory for split PNG assets")
parser.add_argument("--names", help="Comma-separated output names in sorted component order")
parser.add_argument("--sort", choices=["x", "y", "area"], default="x")
parser.add_argument("--threshold", type=int, default=20, help="Alpha threshold for foreground pixels")
parser.add_argument("--min-area", type=int, default=1000, help="Minimum connected area to keep")
parser.add_argument("--connectivity", type=int, choices=[4, 8], default=8)
parser.add_argument("--close-radius", type=int, default=3, help="Small morphology close radius for broken strokes")
parser.add_argument("--merge-gap", type=int, default=18, help="Merge nearby fragments within this pixel gap; 0 disables")
parser.add_argument("--merge-union-growth", type=float, default=2.4, help="Maximum union-box growth allowed when merging fragments")
parser.add_argument("--pad", type=int, default=24)
parser.add_argument("--limit", type=int, help="Maximum number of components to write")
parser.add_argument("--square", action="store_true", help="Place each extracted asset on a transparent square canvas")
parser.add_argument("--manifest", help="Optional JSON report path for component boxes and outputs")
parser.add_argument("--contact-sheet", help="Optional contact sheet image for visual QA")
args = parser.parse_args()
src = Path(args.input)
out_dir = Path(args.out_dir)
out_dir.mkdir(parents=True, exist_ok=True)
image = Image.open(src).convert("RGBA")
mask = foreground_mask(image.getchannel("A"), args.threshold, args.close_radius)
components = component_boxes(mask, args.min_area, args.connectivity)
components = merge_fragments(components, args.merge_gap, args.merge_union_growth)
components = sort_components(components, args.sort)
if args.limit:
components = components[: args.limit]
names = parse_names(args.names)
if names and len(names) != len(components):
raise SystemExit(f"--names supplied {len(names)} names, but detected {len(components)} components")
outputs = []
contact_items = []
for index, component in enumerate(components, start=1):
name = names[index - 1] if names else f"asset_{index:02d}.png"
if not name.lower().endswith(".png"):
name += ".png"
component_image, padded_box = extract_component_asset(image, component["box"], args.pad, args.square)
out_path = out_dir / name
component_image.save(out_path)
entry = {
"path": str(out_path),
"source": str(src),
"box": component["box"],
"padded_box": padded_box,
"area": component["area"],
"merged_count": component.get("merged_count", 1),
"size": list(component_image.size),
}
outputs.append(entry)
contact_items.append({"name": name, "image": component_image})
print(f"{name}: box={component['box']} area={component['area']} size={component_image.size}")
if args.manifest:
Path(args.manifest).write_text(json.dumps({"source": str(src), "assets": outputs}, ensure_ascii=False, indent=2))
if args.contact_sheet:
write_contact_sheet(contact_items, Path(args.contact_sheet))
if __name__ == "__main__":
main()
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| canvas-designanthropics/skillsCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| docxanthropics/skillsUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pdfanthropics/skillsUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| pptxanthropics/skillsUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |