265c61df1a
replaced by blank images in the stitching process and completly broken images are also replaced by blank images which are passed through and are classified as 'other' with all_pased == False. This should be changed in the future by introducing a new key to the response, indicating that the image is not valid.
81 lines
3.0 KiB
Python
81 lines
3.0 KiB
Python
import math
|
|
from operator import itemgetter
|
|
|
|
from image_prediction.config import CONFIG
|
|
from image_prediction.transformer.transformer import Transformer
|
|
from image_prediction.utils import get_logger
|
|
|
|
logger = get_logger()
|
|
|
|
|
|
class ResponseTransformer(Transformer):
|
|
def transform(self, data):
|
|
logger.debug("ResponseTransformer.transform")
|
|
return build_image_info(data)
|
|
|
|
|
|
def build_image_info(data: dict) -> dict:
|
|
def compute_geometric_quotient():
|
|
page_area_sqrt = math.sqrt(abs(page_width * page_height))
|
|
image_area_sqrt = math.sqrt(abs(x2 - x1) * abs(y2 - y1))
|
|
return image_area_sqrt / page_area_sqrt
|
|
|
|
page_width, page_height, x1, x2, y1, y2, width, height, alpha = itemgetter(
|
|
"page_width", "page_height", "x1", "x2", "y1", "y2", "width", "height", "alpha"
|
|
)(data)
|
|
|
|
quotient = round(compute_geometric_quotient(), 4)
|
|
|
|
min_image_to_page_quotient_breached = bool(quotient < CONFIG.filters.image_to_page_quotient.min)
|
|
max_image_to_page_quotient_breached = bool(quotient > CONFIG.filters.image_to_page_quotient.max)
|
|
min_image_width_to_height_quotient_breached = bool(
|
|
width / height < CONFIG.filters.image_width_to_height_quotient.min
|
|
)
|
|
max_image_width_to_height_quotient_breached = bool(
|
|
width / height > CONFIG.filters.image_width_to_height_quotient.max
|
|
)
|
|
|
|
# FIXME: pass in fallback value for classification and introduce new key for image validness
|
|
classification = data["classification"] or "other"
|
|
representation = data["representation"]
|
|
|
|
min_confidence_breached = (
|
|
bool(max(classification["probabilities"].values()) < CONFIG.filters.min_confidence)
|
|
if data["classification"]
|
|
else True
|
|
)
|
|
|
|
image_info = {
|
|
"classification": classification,
|
|
"representation": representation,
|
|
"position": {"x1": x1, "x2": x2, "y1": y1, "y2": y2, "pageNumber": data["page_idx"] + 1},
|
|
"geometry": {"width": width, "height": height},
|
|
"alpha": alpha,
|
|
"filters": {
|
|
"geometry": {
|
|
"imageSize": {
|
|
"quotient": quotient,
|
|
"tooLarge": max_image_to_page_quotient_breached,
|
|
"tooSmall": min_image_to_page_quotient_breached,
|
|
},
|
|
"imageFormat": {
|
|
"quotient": round(width / height, 4),
|
|
"tooTall": min_image_width_to_height_quotient_breached,
|
|
"tooWide": max_image_width_to_height_quotient_breached,
|
|
},
|
|
},
|
|
"probability": {"unconfident": min_confidence_breached},
|
|
"allPassed": not any(
|
|
[
|
|
max_image_to_page_quotient_breached,
|
|
min_image_to_page_quotient_breached,
|
|
min_image_width_to_height_quotient_breached,
|
|
max_image_width_to_height_quotient_breached,
|
|
min_confidence_breached,
|
|
]
|
|
),
|
|
},
|
|
}
|
|
|
|
return image_info
|