refactor: remove needless module

src/arcaea_offline_ocr/__init__.py

@@ -1,5 +1,4 @@
 from .crop import *
 from .device import *
-from .mask import *
 from .ocr import *
 from .utils import *

src/arcaea_offline_ocr/b30/chieri/v4/ocr.py

@@ -8,7 +8,6 @@ from PIL import Image
 from ....crop import crop_xywh
 from ....ocr import FixRects, ocr_digits_by_contour_knn, preprocess_hog
 from ....phash_db import ImagePHashDatabase
-from ....sift_db import SIFTDatabase
 from ....types import Mat, cv2_ml_KNearest
 from ....utils import construct_int_xywh_rect
 from ...shared import B30OcrResultItem

src/arcaea_offline_ocr/device/ocr.py

@@ -0,0 +1,101 @@
+import cv2
+import numpy as np
+from PIL import Image
+
+from ..crop import crop_xywh
+from ..ocr import (
+    FixRects,
+    ocr_digit_samples_knn,
+    ocr_digits_by_contour_knn,
+    preprocess_hog,
+    resize_fill_square,
+)
+from ..phash_db import ImagePHashDatabase
+from .roi.extractor import DeviceRoiExtractor
+from .roi.masker import DeviceRoiMasker
+
+
+class DeviceOcr:
+    def __init__(
+        self,
+        extractor: DeviceRoiExtractor,
+        masker: DeviceRoiMasker,
+        knn_model: cv2.ml.KNearest,
+        phash_db: ImagePHashDatabase,
+    ):
+        self.extractor = extractor
+        self.masker = masker
+        self.knn_model = knn_model
+        self.phash_db = phash_db
+
+    def pfl(self, roi_gray: cv2.Mat, factor: float = 1.25):
+        contours, _ = cv2.findContours(
+            roi_gray, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE
+        )
+        filtered_contours = [c for c in contours if cv2.contourArea(c) >= 5 * factor]
+        rects = [cv2.boundingRect(c) for c in filtered_contours]
+        rects = FixRects.connect_broken(rects, roi_gray.shape[1], roi_gray.shape[0])
+
+        filtered_rects = [r for r in rects if r[2] >= 5 * factor and r[3] >= 6 * factor]
+        filtered_rects = FixRects.split_connected(roi_gray, filtered_rects)
+        filtered_rects = sorted(filtered_rects, key=lambda r: r[0])
+
+        roi_ocr = roi_gray.copy()
+        filtered_contours_flattened = {tuple(c.flatten()) for c in filtered_contours}
+        for contour in contours:
+            if tuple(contour.flatten()) in filtered_contours_flattened:
+                continue
+            roi_ocr = cv2.fillPoly(roi_ocr, [contour], [0])
+        digit_rois = [
+            resize_fill_square(crop_xywh(roi_ocr, r), 20)
+            for r in sorted(filtered_rects, key=lambda r: r[0])
+        ]
+
+        samples = preprocess_hog(digit_rois)
+        return ocr_digit_samples_knn(samples, self.knn_model)
+
+    def pure(self):
+        return self.pfl(self.masker.pure(self.extractor.pure))
+
+    def far(self):
+        return self.pfl(self.masker.far(self.extractor.far))
+
+    def lost(self):
+        return self.pfl(self.masker.lost(self.extractor.lost))
+
+    def score(self):
+        roi = self.masker.score(self.extractor.score)
+        contours, _ = cv2.findContours(roi, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
+        for contour in contours:
+            x, y, w, h = cv2.boundingRect(contour)
+            if h < roi.shape[0] * 0.6:
+                roi = cv2.fillPoly(roi, [contour], [0])
+        return ocr_digits_by_contour_knn(roi, self.knn_model)
+
+    def rating_class(self):
+        roi = self.extractor.rating_class
+        results = [
+            self.masker.rating_class_pst(roi),
+            self.masker.rating_class_prs(roi),
+            self.masker.rating_class_ftr(roi),
+            self.masker.rating_class_byd(roi),
+        ]
+        return max(enumerate(results), key=lambda i: np.count_nonzero(i[1]))[0]
+
+    def max_recall(self):
+        return ocr_digits_by_contour_knn(
+            self.masker.max_recall(self.extractor.max_recall), self.knn_model
+        )
+
+    def clear_status(self):
+        roi = self.extractor.clear_status
+        results = [
+            self.masker.clear_status_track_lost(roi),
+            self.masker.clear_status_track_complete(roi),
+            self.masker.clear_status_full_recall(roi),
+            self.masker.clear_status_pure_memory(roi),
+        ]
+        return max(enumerate(results), key=lambda i: np.count_nonzero(i[1]))[0]
+
+    def song_id(self):
+        return self.phash_db.lookup_image(Image.fromarray(self.extractor.jacket))[0]

src/arcaea_offline_ocr/device/roi/definitions/auto/__init__.py

@@ -0,0 +1,3 @@
+from .common import DeviceAutoRoiSizes
+from .t1 import DeviceAutoRoiSizesT1
+from .t2 import DeviceAutoRoiSizesT2

src/arcaea_offline_ocr/device/roi/definitions/auto/common.py

@@ -1,7 +1,7 @@
-from ..common import Sizes
+from ..common import DeviceRoiSizes
 
 
-class AutoSizes(Sizes):
+class DeviceAutoRoiSizes(DeviceRoiSizes):
     def __init__(self, w: int, h: int):
         self.w = w
         self.h = h

src/arcaea_offline_ocr/device/roi/definitions/auto/t1.py

@@ -1,7 +1,7 @@
-from .common import AutoSizes
+from .common import DeviceAutoRoiSizes
 
 
-class AutoSizesT1(AutoSizes):
+class DeviceAutoRoiSizesT1(DeviceAutoRoiSizes):
     @property
     def factor(self):
         return (

src/arcaea_offline_ocr/device/roi/definitions/auto/t2.py

@@ -1,7 +1,7 @@
-from .common import AutoSizes
+from .common import DeviceAutoRoiSizes
 
 
-class AutoSizesT2(AutoSizes):
+class DeviceAutoRoiSizesT2(DeviceAutoRoiSizes):
     @property
     def factor(self):
         return (

src/arcaea_offline_ocr/device/roi/definitions/common.py

@@ -3,7 +3,7 @@ from typing import Tuple
 Rect = Tuple[int, int, int, int]
 
 
-class Sizes:
+class DeviceRoiSizes:
     pure: Rect
     far: Rect
     lost: Rect

src/arcaea_offline_ocr/device/roi/extractor/__init__.py

@@ -0,0 +1 @@
+from .common import DeviceRoiExtractor

src/arcaea_offline_ocr/device/roi/extractor/common.py

@@ -1,11 +1,11 @@
 import cv2
 
-from ..crop import crop_xywh
-from .sizes.common import Sizes
+from ....crop import crop_xywh
+from ..definitions.common import DeviceRoiSizes
 
 
-class Extractor:
-    def __init__(self, img: cv2.Mat, sizes: Sizes):
+class DeviceRoiExtractor:
+    def __init__(self, img: cv2.Mat, sizes: DeviceRoiSizes):
         self.img = img
         self.sizes = sizes
 
@@ -28,6 +28,10 @@ class Extractor:
     def score(self):
         return crop_xywh(self.img, self.__construct_int_rect(self.sizes.score))
 
+    @property
+    def jacket(self):
+        return crop_xywh(self.img, self.__construct_int_rect(self.sizes.jacket))
+
     @property
     def rating_class(self):
         return crop_xywh(self.img, self.__construct_int_rect(self.sizes.rating_class))

src/arcaea_offline_ocr/device/roi/masker/__init__.py

@@ -0,0 +1,2 @@
+from .auto import *
+from .common import DeviceRoiMasker

src/arcaea_offline_ocr/device/roi/masker/auto/__init__.py

@@ -0,0 +1,3 @@
+from .common import DeviceAutoRoiMasker
+from .t1 import DeviceAutoRoiMaskerT1
+from .t2 import DeviceAutoRoiMaskerT2

src/arcaea_offline_ocr/device/roi/masker/auto/common.py

@@ -0,0 +1,5 @@
+from ..common import DeviceRoiMasker
+
+
+class DeviceAutoRoiMasker(DeviceRoiMasker):
+    ...

src/arcaea_offline_ocr/device/roi/masker/auto/t1.py

@@ -0,0 +1,123 @@
+import cv2
+import numpy as np
+
+from .common import DeviceAutoRoiMasker
+
+GRAY_BGR_MIN = np.array([50] * 3, np.uint8)
+GRAY_BGR_MAX = np.array([160] * 3, np.uint8)
+
+WHITE_HSV_MIN = np.array([0, 0, 240], np.uint8)
+WHITE_HSV_MAX = np.array([179, 10, 255], np.uint8)
+
+PST_HSV_MIN = np.array([100, 50, 80], np.uint8)
+PST_HSV_MAX = np.array([100, 255, 255], np.uint8)
+
+PRS_HSV_MIN = np.array([43, 40, 75], np.uint8)
+PRS_HSV_MAX = np.array([50, 155, 190], np.uint8)
+
+FTR_HSV_MIN = np.array([149, 30, 0], np.uint8)
+FTR_HSV_MAX = np.array([155, 181, 150], np.uint8)
+
+BYD_HSV_MIN = np.array([170, 50, 50], np.uint8)
+BYD_HSV_MAX = np.array([179, 210, 198], np.uint8)
+
+TRACK_LOST_HSV_MIN = np.array([170, 75, 90], np.uint8)
+TRACK_LOST_HSV_MAX = np.array([175, 170, 160], np.uint8)
+
+TRACK_COMPLETE_HSV_MIN = np.array([140, 0, 50], np.uint8)
+TRACK_COMPLETE_HSV_MAX = np.array([145, 50, 130], np.uint8)
+
+FULL_RECALL_HSV_MIN = np.array([140, 60, 80], np.uint8)
+FULL_RECALL_HSV_MAX = np.array([150, 130, 145], np.uint8)
+
+PURE_MEMORY_HSV_MIN = np.array([90, 70, 80], np.uint8)
+PURE_MEMORY_HSV_MAX = np.array([110, 200, 175], np.uint8)
+
+
+class DeviceAutoRoiMaskerT1(DeviceAutoRoiMasker):
+    @classmethod
+    def gray(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        bgr_value_equal_mask = np.max(roi_bgr, axis=2) - np.min(roi_bgr, axis=2) <= 5
+        img_bgr = roi_bgr.copy()
+        img_bgr[~bgr_value_equal_mask] = np.array([0, 0, 0], roi_bgr.dtype)
+        img_bgr = cv2.erode(img_bgr, cv2.getStructuringElement(cv2.MORPH_RECT, (2, 2)))
+        img_bgr = cv2.dilate(img_bgr, cv2.getStructuringElement(cv2.MORPH_RECT, (1, 1)))
+        return cv2.inRange(img_bgr, GRAY_BGR_MIN, GRAY_BGR_MAX)
+
+    @classmethod
+    def pure(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cls.gray(roi_bgr)
+
+    @classmethod
+    def far(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cls.gray(roi_bgr)
+
+    @classmethod
+    def lost(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cls.gray(roi_bgr)
+
+    @classmethod
+    def score(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), WHITE_HSV_MIN, WHITE_HSV_MAX
+        )
+
+    @classmethod
+    def rating_class_pst(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), PST_HSV_MIN, PST_HSV_MAX
+        )
+
+    @classmethod
+    def rating_class_prs(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), PRS_HSV_MIN, PRS_HSV_MAX
+        )
+
+    @classmethod
+    def rating_class_ftr(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), FTR_HSV_MIN, FTR_HSV_MAX
+        )
+
+    @classmethod
+    def rating_class_byd(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), BYD_HSV_MIN, BYD_HSV_MAX
+        )
+
+    @classmethod
+    def max_recall(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cls.gray(roi_bgr)
+
+    @classmethod
+    def clear_status_track_lost(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            TRACK_LOST_HSV_MIN,
+            TRACK_LOST_HSV_MAX,
+        )
+
+    @classmethod
+    def clear_status_track_complete(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            TRACK_COMPLETE_HSV_MIN,
+            TRACK_COMPLETE_HSV_MAX,
+        )
+
+    @classmethod
+    def clear_status_full_recall(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            FULL_RECALL_HSV_MIN,
+            FULL_RECALL_HSV_MAX,
+        )
+
+    @classmethod
+    def clear_status_pure_memory(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            PURE_MEMORY_HSV_MIN,
+            PURE_MEMORY_HSV_MAX,
+        )

src/arcaea_offline_ocr/device/roi/masker/auto/t2.py

@@ -0,0 +1,128 @@
+import cv2
+import numpy as np
+
+from .common import DeviceAutoRoiMasker
+
+PFL_HSV_MIN = np.array([0, 0, 248], np.uint8)
+PFL_HSV_MAX = np.array([179, 10, 255], np.uint8)
+
+WHITE_HSV_MIN = np.array([0, 0, 240], np.uint8)
+WHITE_HSV_MAX = np.array([179, 10, 255], np.uint8)
+
+
+PST_HSV_MIN = np.array([100, 50, 80], np.uint8)
+PST_HSV_MAX = np.array([100, 255, 255], np.uint8)
+
+PRS_HSV_MIN = np.array([43, 40, 75], np.uint8)
+PRS_HSV_MAX = np.array([50, 155, 190], np.uint8)
+
+FTR_HSV_MIN = np.array([149, 30, 0], np.uint8)
+FTR_HSV_MAX = np.array([155, 181, 150], np.uint8)
+
+BYD_HSV_MIN = np.array([170, 50, 50], np.uint8)
+BYD_HSV_MAX = np.array([179, 210, 198], np.uint8)
+
+MAX_RECALL_HSV_MIN = np.array([125, 0, 0], np.uint8)
+MAX_RECALL_HSV_MAX = np.array([130, 100, 150], np.uint8)
+
+TRACK_LOST_HSV_MIN = np.array([170, 75, 90], np.uint8)
+TRACK_LOST_HSV_MAX = np.array([175, 170, 160], np.uint8)
+
+TRACK_COMPLETE_HSV_MIN = np.array([140, 0, 50], np.uint8)
+TRACK_COMPLETE_HSV_MAX = np.array([145, 50, 130], np.uint8)
+
+FULL_RECALL_HSV_MIN = np.array([140, 60, 80], np.uint8)
+FULL_RECALL_HSV_MAX = np.array([150, 130, 145], np.uint8)
+
+PURE_MEMORY_HSV_MIN = np.array([90, 70, 80], np.uint8)
+PURE_MEMORY_HSV_MAX = np.array([110, 200, 175], np.uint8)
+
+
+class DeviceAutoRoiMaskerT2(DeviceAutoRoiMasker):
+    @classmethod
+    def pfl(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), PFL_HSV_MIN, PFL_HSV_MAX
+        )
+
+    @classmethod
+    def pure(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cls.pfl(roi_bgr)
+
+    @classmethod
+    def far(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cls.pfl(roi_bgr)
+
+    @classmethod
+    def lost(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cls.pfl(roi_bgr)
+
+    @classmethod
+    def score(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), WHITE_HSV_MIN, WHITE_HSV_MAX
+        )
+
+    @classmethod
+    def rating_class_pst(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), PST_HSV_MIN, PST_HSV_MAX
+        )
+
+    @classmethod
+    def rating_class_prs(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), PRS_HSV_MIN, PRS_HSV_MAX
+        )
+
+    @classmethod
+    def rating_class_ftr(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), FTR_HSV_MIN, FTR_HSV_MAX
+        )
+
+    @classmethod
+    def rating_class_byd(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV), BYD_HSV_MIN, BYD_HSV_MAX
+        )
+
+    @classmethod
+    def max_recall(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            MAX_RECALL_HSV_MIN,
+            MAX_RECALL_HSV_MAX,
+        )
+
+    @classmethod
+    def clear_status_track_lost(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            TRACK_LOST_HSV_MIN,
+            TRACK_LOST_HSV_MAX,
+        )
+
+    @classmethod
+    def clear_status_track_complete(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            TRACK_COMPLETE_HSV_MIN,
+            TRACK_COMPLETE_HSV_MAX,
+        )
+
+    @classmethod
+    def clear_status_full_recall(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            FULL_RECALL_HSV_MIN,
+            FULL_RECALL_HSV_MAX,
+        )
+
+    @classmethod
+    def clear_status_pure_memory(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        return cv2.inRange(
+            cv2.cvtColor(roi_bgr, cv2.COLOR_BGR2HSV),
+            PURE_MEMORY_HSV_MIN,
+            PURE_MEMORY_HSV_MAX,
+        )

src/arcaea_offline_ocr/device/roi/masker/common.py

@@ -0,0 +1,55 @@
+import cv2
+
+
+class DeviceRoiMasker:
+    @classmethod
+    def pure(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def far(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def lost(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def score(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def rating_class_pst(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def rating_class_prs(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def rating_class_ftr(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def rating_class_byd(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def max_recall(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def clear_status_track_lost(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def clear_status_track_complete(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def clear_status_full_recall(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()
+
+    @classmethod
+    def clear_status_pure_memory(cls, roi_bgr: cv2.Mat) -> cv2.Mat:
+        raise NotImplementedError()

src/arcaea_offline_ocr/device/v1/crop.py

@@ -1,53 +0,0 @@
-from typing import Tuple
-
-from ...types import Mat
-from .definition import DeviceV1
-
-__all__ = [
-    "crop_img",
-    "crop_from_device_attr",
-    "crop_to_pure",
-    "crop_to_far",
-    "crop_to_lost",
-    "crop_to_max_recall",
-    "crop_to_rating_class",
-    "crop_to_score",
-    "crop_to_title",
-]
-
-
-def crop_img(img: Mat, *, top: int, left: int, bottom: int, right: int):
-    return img[top:bottom, left:right]
-
-
-def crop_from_device_attr(img: Mat, rect: Tuple[int, int, int, int]):
-    x, y, w, h = rect
-    return crop_img(img, top=y, left=x, bottom=y + h, right=x + w)
-
-
-def crop_to_pure(screenshot: Mat, device: DeviceV1):
-    return crop_from_device_attr(screenshot, device.pure)
-
-
-def crop_to_far(screenshot: Mat, device: DeviceV1):
-    return crop_from_device_attr(screenshot, device.far)
-
-
-def crop_to_lost(screenshot: Mat, device: DeviceV1):
-    return crop_from_device_attr(screenshot, device.lost)
-
-
-def crop_to_max_recall(screenshot: Mat, device: DeviceV1):
-    return crop_from_device_attr(screenshot, device.max_recall)
-
-
-def crop_to_rating_class(screenshot: Mat, device: DeviceV1):
-    return crop_from_device_attr(screenshot, device.rating_class)
-
-
-def crop_to_score(screenshot: Mat, device: DeviceV1):
-    return crop_from_device_attr(screenshot, device.score)
-
-
-def crop_to_title(screenshot: Mat, device: DeviceV1):
-    return crop_from_device_attr(screenshot, device.title)

src/arcaea_offline_ocr/device/v1/definition.py

@@ -1,37 +0,0 @@
-from dataclasses import dataclass
-from typing import Any, Dict, Tuple
-
-__all__ = ["DeviceV1"]
-
-
-@dataclass(kw_only=True)
-class DeviceV1:
-    version: int
-    uuid: str
-    name: str
-    pure: Tuple[int, int, int, int]
-    far: Tuple[int, int, int, int]
-    lost: Tuple[int, int, int, int]
-    max_recall: Tuple[int, int, int, int]
-    rating_class: Tuple[int, int, int, int]
-    score: Tuple[int, int, int, int]
-    title: Tuple[int, int, int, int]
-
-    @classmethod
-    def from_json_object(cls, json_dict: Dict[str, Any]):
-        if json_dict["version"] == 1:
-            return cls(
-                version=1,
-                uuid=json_dict["uuid"],
-                name=json_dict["name"],
-                pure=json_dict["pure"],
-                far=json_dict["far"],
-                lost=json_dict["lost"],
-                max_recall=json_dict["max_recall"],
-                rating_class=json_dict["rating_class"],
-                score=json_dict["score"],
-                title=json_dict["title"],
-            )
-
-    def repr_info(self):
-        return f"Device(version={self.version}, uuid={repr(self.uuid)}, name={repr(self.name)})"

src/arcaea_offline_ocr/device/v1/ocr.py

@@ -1,86 +0,0 @@
-from typing import List
-
-import cv2
-
-from ...crop import crop_xywh
-from ...mask import mask_gray, mask_white
-from ...ocr import ocr_digits_by_contour_knn, ocr_rating_class
-from ...types import Mat, cv2_ml_KNearest
-from ..shared import DeviceOcrResult
-from .crop import *
-from .definition import DeviceV1
-
-
-class DeviceV1Ocr:
-    def __init__(self, device: DeviceV1, knn_model: cv2_ml_KNearest):
-        self.__device = device
-        self.__knn_model = knn_model
-
-    @property
-    def device(self):
-        return self.__device
-
-    @device.setter
-    def device(self, value):
-        self.__device = value
-
-    @property
-    def knn_model(self):
-        return self.__knn_model
-
-    @knn_model.setter
-    def knn_model(self, value):
-        self.__knn_model = value
-
-    def preprocess_score_roi(self, __roi_gray: Mat) -> List[Mat]:
-        roi_gray = __roi_gray.copy()
-        contours, _ = cv2.findContours(
-            roi_gray, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE
-        )
-        for contour in contours:
-            rect = cv2.boundingRect(contour)
-            if rect[3] > roi_gray.shape[0] * 0.6:
-                continue
-            roi_gray = cv2.fillPoly(roi_gray, [contour], 0)
-        return roi_gray
-
-    def ocr(self, img_bgr: Mat):
-        rating_class_roi = crop_to_rating_class(img_bgr, self.device)
-        rating_class = ocr_rating_class(rating_class_roi)
-
-        pfl_mr_roi = [
-            crop_to_pure(img_bgr, self.device),
-            crop_to_far(img_bgr, self.device),
-            crop_to_lost(img_bgr, self.device),
-            crop_to_max_recall(img_bgr, self.device),
-        ]
-        pfl_mr_roi = [mask_gray(roi) for roi in pfl_mr_roi]
-
-        pure, far, lost = [
-            ocr_digits_by_contour_knn(roi, self.knn_model) for roi in pfl_mr_roi[:3]
-        ]
-
-        max_recall_contours, _ = cv2.findContours(
-            pfl_mr_roi[3], cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE
-        )
-        max_recall_rects = [cv2.boundingRect(c) for c in max_recall_contours]
-        max_recall_rect = sorted(max_recall_rects, key=lambda r: r[0])[-1]
-        max_recall_roi = crop_xywh(img_bgr, max_recall_rect)
-        max_recall = ocr_digits_by_contour_knn(max_recall_roi, self.knn_model)
-
-        score_roi = crop_to_score(img_bgr, self.device)
-        score_roi = mask_white(score_roi)
-        score_roi = self.preprocess_score_roi(score_roi)
-        score = ocr_digits_by_contour_knn(score_roi, self.knn_model)
-
-        return DeviceOcrResult(
-            song_id=None,
-            title=None,
-            rating_class=rating_class,
-            pure=pure,
-            far=far,
-            lost=lost,
-            score=score,
-            max_recall=max_recall,
-            clear_type=None,
-        )

src/arcaea_offline_ocr/device/v2/__init__.py

@@ -1,4 +0,0 @@
-from .definition import DeviceV2
-from .ocr import DeviceV2Ocr
-from .rois import DeviceV2AutoRois, DeviceV2Rois
-from .shared import MAX_RECALL_CLOSE_KERNEL

src/arcaea_offline_ocr/device/v2/definition.py

@@ -1,26 +0,0 @@
-from typing import Iterable
-
-from attrs import define, field
-
-from ...types import XYWHRect
-
-
-def iterable_to_xywh_rect(__iter: Iterable) -> XYWHRect:
-    return XYWHRect(*__iter)
-
-
-@define(kw_only=True)
-class DeviceV2:
-    version = field(type=int)
-    uuid = field(type=str)
-    name = field(type=str)
-    crop_black_edges = field(type=bool)
-    factor = field(type=float)
-    pure = field(converter=iterable_to_xywh_rect, default=[0, 0, 0, 0])
-    far = field(converter=iterable_to_xywh_rect, default=[0, 0, 0, 0])
-    lost = field(converter=iterable_to_xywh_rect, default=[0, 0, 0, 0])
-    score = field(converter=iterable_to_xywh_rect, default=[0, 0, 0, 0])
-    max_recall_rating_class = field(
-        converter=iterable_to_xywh_rect, default=[0, 0, 0, 0]
-    )
-    title = field(converter=iterable_to_xywh_rect, default=[0, 0, 0, 0])

src/arcaea_offline_ocr/device/v2/ocr.py

@@ -1,172 +0,0 @@
-import math
-from functools import lru_cache
-from typing import Sequence
-
-import cv2
-import numpy as np
-from PIL import Image
-
-from ...crop import crop_xywh
-from ...mask import (
-    mask_byd,
-    mask_ftr,
-    mask_gray,
-    mask_max_recall_purple,
-    mask_pfl_white,
-    mask_prs,
-    mask_pst,
-    mask_white,
-)
-from ...ocr import (
-    FixRects,
-    ocr_digit_samples_knn,
-    ocr_digits_by_contour_knn,
-    preprocess_hog,
-    resize_fill_square,
-)
-from ...phash_db import ImagePHashDatabase
-from ...sift_db import SIFTDatabase
-from ...types import Mat, cv2_ml_KNearest
-from ..shared import DeviceOcrResult
-from .preprocess import find_digits_preprocess
-from .rois import DeviceV2Rois
-from .shared import MAX_RECALL_CLOSE_KERNEL
-from .sizes import SizesV2
-
-
-class DeviceV2Ocr:
-    def __init__(self, knn_model: cv2_ml_KNearest, phash_db: ImagePHashDatabase):
-        self.__knn_model = knn_model
-        self.__phash_db = phash_db
-
-    @property
-    def knn_model(self):
-        if not self.__knn_model:
-            raise ValueError("`knn_model` unset.")
-        return self.__knn_model
-
-    @knn_model.setter
-    def knn_model(self, value: cv2_ml_KNearest):
-        self.__knn_model = value
-
-    @property
-    def phash_db(self):
-        if not self.__phash_db:
-            raise ValueError("`phash_db` unset.")
-        return self.__phash_db
-
-    @phash_db.setter
-    def phash_db(self, value: SIFTDatabase):
-        self.__phash_db = value
-
-    @lru_cache
-    def _get_digit_widths(self, num_list: Sequence[int], factor: float):
-        widths = set()
-        for n in num_list:
-            lower = math.floor(n * factor)
-            upper = math.ceil(n * factor)
-            widths.update(range(lower, upper + 1))
-        return widths
-
-    def _base_ocr_pfl(self, roi_masked: Mat, factor: float = 1.0):
-        contours, _ = cv2.findContours(
-            roi_masked, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE
-        )
-        filtered_contours = [c for c in contours if cv2.contourArea(c) >= 5 * factor]
-        rects = [cv2.boundingRect(c) for c in filtered_contours]
-        rects = FixRects.connect_broken(rects, roi_masked.shape[1], roi_masked.shape[0])
-        rect_contour_map = dict(zip(rects, filtered_contours))
-
-        filtered_rects = [r for r in rects if r[2] >= 5 * factor and r[3] >= 6 * factor]
-        filtered_rects = FixRects.split_connected(roi_masked, filtered_rects)
-        filtered_rects = sorted(filtered_rects, key=lambda r: r[0])
-
-        roi_ocr = roi_masked.copy()
-        filtered_contours_flattened = {tuple(c.flatten()) for c in filtered_contours}
-        for contour in contours:
-            if tuple(contour.flatten()) in filtered_contours_flattened:
-                continue
-            roi_ocr = cv2.fillPoly(roi_ocr, [contour], [0])
-        digit_rois = [
-            resize_fill_square(crop_xywh(roi_ocr, r), 20)
-            for r in sorted(filtered_rects, key=lambda r: r[0])
-        ]
-        # [cv2.imshow(f"r{i}", r) for i, r in enumerate(digit_rois)]
-        # cv2.waitKey(0)
-        samples = preprocess_hog(digit_rois)
-        return ocr_digit_samples_knn(samples, self.knn_model)
-
-    def ocr_song_id(self, rois: DeviceV2Rois):
-        jacket = cv2.cvtColor(rois.jacket, cv2.COLOR_BGR2GRAY)
-        return self.phash_db.lookup_image(Image.fromarray(jacket))[0]
-
-    def ocr_rating_class(self, rois: DeviceV2Rois):
-        roi = cv2.cvtColor(rois.max_recall_rating_class, cv2.COLOR_BGR2HSV)
-        results = [mask_pst(roi), mask_prs(roi), mask_ftr(roi), mask_byd(roi)]
-        return max(enumerate(results), key=lambda i: np.count_nonzero(i[1]))[0]
-
-    def ocr_score(self, rois: DeviceV2Rois):
-        roi = cv2.cvtColor(rois.score, cv2.COLOR_BGR2HSV)
-        roi = mask_white(roi)
-        contours, _ = cv2.findContours(roi, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
-        for contour in contours:
-            x, y, w, h = cv2.boundingRect(contour)
-            if h < roi.shape[0] * 0.6:
-                roi = cv2.fillPoly(roi, [contour], [0])
-        return ocr_digits_by_contour_knn(roi, self.knn_model)
-
-    def mask_pfl(self, pfl_roi: Mat, rois: DeviceV2Rois):
-        return (
-            mask_pfl_white(cv2.cvtColor(pfl_roi, cv2.COLOR_BGR2HSV))
-            if isinstance(rois.sizes, SizesV2)
-            else mask_gray(pfl_roi)
-        )
-
-    def ocr_pure(self, rois: DeviceV2Rois):
-        roi = self.mask_pfl(rois.pure, rois)
-        return self._base_ocr_pfl(roi, rois.sizes.factor)
-
-    def ocr_far(self, rois: DeviceV2Rois):
-        roi = self.mask_pfl(rois.far, rois)
-        return self._base_ocr_pfl(roi, rois.sizes.factor)
-
-    def ocr_lost(self, rois: DeviceV2Rois):
-        roi = self.mask_pfl(rois.lost, rois)
-        return self._base_ocr_pfl(roi, rois.sizes.factor)
-
-    def ocr_max_recall(self, rois: DeviceV2Rois):
-        roi = (
-            mask_max_recall_purple(
-                cv2.cvtColor(rois.max_recall_rating_class, cv2.COLOR_BGR2HSV)
-            )
-            if isinstance(rois.sizes, SizesV2)
-            else mask_gray(rois.max_recall_rating_class)
-        )
-        roi_closed = cv2.morphologyEx(roi, cv2.MORPH_CLOSE, MAX_RECALL_CLOSE_KERNEL)
-        contours, _ = cv2.findContours(
-            roi_closed, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE
-        )
-        rects = sorted(
-            [cv2.boundingRect(c) for c in contours], key=lambda r: r[0], reverse=True
-        )
-        max_recall_roi = crop_xywh(roi, rects[0])
-        return ocr_digits_by_contour_knn(max_recall_roi, self.knn_model)
-
-    def ocr(self, rois: DeviceV2Rois):
-        song_id = self.ocr_song_id(rois)
-        rating_class = self.ocr_rating_class(rois)
-        score = self.ocr_score(rois)
-        pure = self.ocr_pure(rois)
-        far = self.ocr_far(rois)
-        lost = self.ocr_lost(rois)
-        max_recall = self.ocr_max_recall(rois)
-
-        return DeviceOcrResult(
-            rating_class=rating_class,
-            pure=pure,
-            far=far,
-            lost=lost,
-            score=score,
-            max_recall=max_recall,
-            song_id=song_id,
-        )

src/arcaea_offline_ocr/device/v2/preprocess.py

@@ -1,54 +0,0 @@
-import cv2
-
-from ...types import Mat
-from .shared import *
-
-
-def find_digits_preprocess(__img_masked: Mat) -> Mat:
-    img = __img_masked.copy()
-    img_denoised = cv2.morphologyEx(img, cv2.MORPH_OPEN, PFL_DENOISE_KERNEL)
-    # img_denoised = cv2.bitwise_and(img, img_denoised)
-
-    denoise_contours, _ = cv2.findContours(
-        img_denoised, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE
-    )
-    # cv2.drawContours(img_denoised, contours, -1, [128], 2)
-
-    # fill all contour.area < max(contour.area) * ratio with black pixels
-    # for denoise purposes
-
-    # define threshold contour area
-    # we assume the smallest digit "1", is 80% height of the image,
-    # and at least 1.5 pixel wide, considering cv2.contourArea always
-    # returns a smaller value than the actual contour area.
-    max_contour_area = __img_masked.shape[0] * 0.8 * 1.5
-    filtered_contours = list(
-        filter(lambda c: cv2.contourArea(c) >= max_contour_area, denoise_contours)
-    )
-
-    filtered_contours_flattened = {tuple(c.flatten()) for c in filtered_contours}
-
-    for contour in denoise_contours:
-        if tuple(contour.flatten()) not in filtered_contours_flattened:
-            img_denoised = cv2.fillPoly(img_denoised, [contour], [0])
-
-    # old algorithm, finding the largest contour area
-    ## contour_area_tuples = [(contour, cv2.contourArea(contour)) for contour in contours]
-    ## contour_area_tuples = sorted(
-    ##     contour_area_tuples, key=lambda item: item[1], reverse=True
-    ## )
-    ## max_contour_area = contour_area_tuples[0][1]
-    ## print(max_contour_area, [item[1] for item in contour_area_tuples])
-    ## contours_filter_end_index = len(contours)
-    ## for i, item in enumerate(contour_area_tuples):
-    ##     contour, area = item
-    ##     if area < max_contour_area * 0.15:
-    ##         contours_filter_end_index = i
-    ##         break
-    ## contours = [item[0] for item in contour_area_tuples]
-    ## for contour in contours[-contours_filter_end_index - 1:]:
-    ##     img = cv2.fillPoly(img, [contour], [0])
-    ##     img_denoised = cv2.fillPoly(img_denoised, [contour], [0])
-    ## contours = contours[:contours_filter_end_index]
-
-    return img_denoised

src/arcaea_offline_ocr/device/v2/rois.py

@@ -1,199 +0,0 @@
-from typing import Union
-
-from ...crop import crop_black_edges, crop_xywh
-from ...types import Mat, XYWHRect
-from .definition import DeviceV2
-from .sizes import Sizes, SizesV1
-
-
-def to_int(num: Union[int, float]) -> int:
-    return round(num)
-
-
-class DeviceV2Rois:
-    def __init__(self, device: DeviceV2, img: Mat):
-        self.device = device
-        self.sizes = SizesV1(self.device.factor)
-        self.__img = img
-
-    @staticmethod
-    def construct_int_xywh_rect(x, y, w, h) -> XYWHRect:
-        return XYWHRect(*[to_int(item) for item in [x, y, w, h]])
-
-    @property
-    def img(self):
-        return self.__img
-
-    @img.setter
-    def img(self, img: Mat):
-        self.__img = (
-            crop_black_edges(img) if self.device.crop_black_edges else img.copy()
-        )
-
-    @property
-    def h(self):
-        return self.img.shape[0]
-
-    @property
-    def vmid(self):
-        return self.h / 2
-
-    @property
-    def w(self):
-        return self.img.shape[1]
-
-    @property
-    def hmid(self):
-        return self.w / 2
-
-    @property
-    def h_without_top_bar(self):
-        """img_height -= top_bar_height"""
-        return self.h - self.sizes.TOP_BAR_HEIGHT
-
-    @property
-    def h_without_top_bar_mid(self):
-        return self.sizes.TOP_BAR_HEIGHT + self.h_without_top_bar / 2
-
-    @property
-    def pfl_top(self):
-        return self.h_without_top_bar_mid + self.sizes.PFL_TOP_FROM_VMID
-
-    @property
-    def pfl_left(self):
-        return self.hmid + self.sizes.PFL_LEFT_FROM_HMID
-
-    @property
-    def pure_rect(self):
-        return self.construct_int_xywh_rect(
-            x=self.pfl_left,
-            y=self.pfl_top,
-            w=self.sizes.PFL_WIDTH,
-            h=self.sizes.PFL_FONT_PX,
-        )
-
-    @property
-    def pure(self):
-        return crop_xywh(self.img, self.pure_rect)
-
-    @property
-    def far_rect(self):
-        return self.construct_int_xywh_rect(
-            x=self.pfl_left,
-            y=self.pfl_top + self.sizes.PFL_FONT_PX + self.sizes.PURE_FAR_GAP,
-            w=self.sizes.PFL_WIDTH,
-            h=self.sizes.PFL_FONT_PX,
-        )
-
-    @property
-    def far(self):
-        return crop_xywh(self.img, self.far_rect)
-
-    @property
-    def lost_rect(self):
-        return self.construct_int_xywh_rect(
-            x=self.pfl_left,
-            y=(
-                self.pfl_top
-                + self.sizes.PFL_FONT_PX * 2
-                + self.sizes.PURE_FAR_GAP
-                + self.sizes.FAR_LOST_GAP
-            ),
-            w=self.sizes.PFL_WIDTH,
-            h=self.sizes.PFL_FONT_PX,
-        )
-
-    @property
-    def lost(self):
-        return crop_xywh(self.img, self.lost_rect)
-
-    @property
-    def score_rect(self):
-        return self.construct_int_xywh_rect(
-            x=self.hmid - (self.sizes.SCORE_WIDTH / 2),
-            y=(
-                self.h_without_top_bar_mid
-                + self.sizes.SCORE_BOTTOM_FROM_VMID
-                - self.sizes.SCORE_FONT_PX
-            ),
-            w=self.sizes.SCORE_WIDTH,
-            h=self.sizes.SCORE_FONT_PX,
-        )
-
-    @property
-    def score(self):
-        return crop_xywh(self.img, self.score_rect)
-
-    @property
-    def max_recall_rating_class_rect(self):
-        x = (
-            self.hmid
-            + self.sizes.JACKET_RIGHT_FROM_HOR_MID
-            - self.sizes.JACKET_WIDTH
-            - 25 * self.sizes.factor
-        )
-        return self.construct_int_xywh_rect(
-            x=x,
-            y=(
-                self.h_without_top_bar_mid
-                - self.sizes.SCORE_PANEL[1] / 2
-                - self.sizes.MR_RT_HEIGHT
-            ),
-            w=self.sizes.MR_RT_WIDTH,
-            h=self.sizes.MR_RT_HEIGHT,
-        )
-
-    @property
-    def max_recall_rating_class(self):
-        return crop_xywh(self.img, self.max_recall_rating_class_rect)
-
-    @property
-    def title_rect(self):
-        return self.construct_int_xywh_rect(
-            x=0,
-            y=self.h_without_top_bar_mid
-            + self.sizes.TITLE_BOTTOM_FROM_VMID
-            - self.sizes.TITLE_FONT_PX,
-            w=self.hmid + self.sizes.TITLE_WIDTH_RIGHT,
-            h=self.sizes.TITLE_FONT_PX,
-        )
-
-    @property
-    def title(self):
-        return crop_xywh(self.img, self.title_rect)
-
-    @property
-    def jacket_rect(self):
-        return self.construct_int_xywh_rect(
-            x=self.hmid
-            + self.sizes.JACKET_RIGHT_FROM_HOR_MID
-            - self.sizes.JACKET_WIDTH,
-            y=self.h_without_top_bar_mid - self.sizes.SCORE_PANEL[1] / 2,
-            w=self.sizes.JACKET_WIDTH,
-            h=self.sizes.JACKET_WIDTH,
-        )
-
-    @property
-    def jacket(self):
-        return crop_xywh(self.img, self.jacket_rect)
-
-
-class DeviceV2AutoRois(DeviceV2Rois):
-    @staticmethod
-    def get_factor(width: int, height: int):
-        ratio = width / height
-        return ((width / 16) * 9) / 720 if ratio < (16 / 9) else height / 720
-
-    def __init__(self, img: Mat):
-        factor = self.get_factor(img.shape[1], img.shape[0])
-        self.sizes = SizesV1(factor)
-        self.__img = None
-        self.img = img
-
-    @property
-    def img(self):
-        return self.__img
-
-    @img.setter
-    def img(self, img: Mat):
-        self.__img = crop_black_edges(img)

src/arcaea_offline_ocr/device/v2/shared.py

@@ -1,9 +0,0 @@
-from cv2 import MORPH_RECT, getStructuringElement
-
-PFL_DENOISE_KERNEL = getStructuringElement(MORPH_RECT, [2, 2])
-PFL_ERODE_KERNEL = getStructuringElement(MORPH_RECT, [3, 3])
-PFL_CLOSE_HORIZONTAL_KERNEL = getStructuringElement(MORPH_RECT, [10, 1])
-
-MAX_RECALL_DENOISE_KERNEL = getStructuringElement(MORPH_RECT, [3, 3])
-MAX_RECALL_ERODE_KERNEL = getStructuringElement(MORPH_RECT, [2, 2])
-MAX_RECALL_CLOSE_KERNEL = getStructuringElement(MORPH_RECT, [20, 1])

src/arcaea_offline_ocr/device/v2/sizes.py

@@ -1,254 +0,0 @@
-from typing import Tuple, Union
-
-
-def apply_factor(num: Union[int, float], factor: float):
-    return num * factor
-
-
-class Sizes:
-    def __init__(self, factor: float):
-        raise NotImplementedError()
-
-    @property
-    def TOP_BAR_HEIGHT(self):
-        ...
-
-    @property
-    def SCORE_PANEL(self) -> Tuple[int, int]:
-        ...
-
-    @property
-    def PFL_TOP_FROM_VMID(self):
-        ...
-
-    @property
-    def PFL_LEFT_FROM_HMID(self):
-        ...
-
-    @property
-    def PFL_WIDTH(self):
-        ...
-
-    @property
-    def PFL_FONT_PX(self):
-        ...
-
-    @property
-    def PURE_FAR_GAP(self):
-        ...
-
-    @property
-    def FAR_LOST_GAP(self):
-        ...
-
-    @property
-    def SCORE_BOTTOM_FROM_VMID(self):
-        ...
-
-    @property
-    def SCORE_FONT_PX(self):
-        ...
-
-    @property
-    def SCORE_WIDTH(self):
-        ...
-
-    @property
-    def JACKET_RIGHT_FROM_HOR_MID(self):
-        ...
-
-    @property
-    def JACKET_WIDTH(self):
-        ...
-
-    @property
-    def MR_RT_RIGHT_FROM_HMID(self):
-        ...
-
-    @property
-    def MR_RT_WIDTH(self):
-        ...
-
-    @property
-    def MR_RT_HEIGHT(self):
-        ...
-
-    @property
-    def TITLE_BOTTOM_FROM_VMID(self):
-        ...
-
-    @property
-    def TITLE_FONT_PX(self):
-        ...
-
-    @property
-    def TITLE_WIDTH_RIGHT(self):
-        ...
-
-
-class SizesV1(Sizes):
-    def __init__(self, factor: float):
-        self.factor = factor
-
-    def apply_factor(self, num):
-        return apply_factor(num, self.factor)
-
-    @property
-    def TOP_BAR_HEIGHT(self):
-        return self.apply_factor(50)
-
-    @property
-    def SCORE_PANEL(self) -> Tuple[int, int]:
-        return tuple(self.apply_factor(num) for num in [485, 239])
-
-    @property
-    def PFL_TOP_FROM_VMID(self):
-        return self.apply_factor(135)
-
-    @property
-    def PFL_LEFT_FROM_HMID(self):
-        return self.apply_factor(5)
-
-    @property
-    def PFL_WIDTH(self):
-        return self.apply_factor(76)
-
-    @property
-    def PFL_FONT_PX(self):
-        return self.apply_factor(26)
-
-    @property
-    def PURE_FAR_GAP(self):
-        return self.apply_factor(12)
-
-    @property
-    def FAR_LOST_GAP(self):
-        return self.apply_factor(10)
-
-    @property
-    def SCORE_BOTTOM_FROM_VMID(self):
-        return self.apply_factor(-50)
-
-    @property
-    def SCORE_FONT_PX(self):
-        return self.apply_factor(45)
-
-    @property
-    def SCORE_WIDTH(self):
-        return self.apply_factor(280)
-
-    @property
-    def JACKET_RIGHT_FROM_HOR_MID(self):
-        return self.apply_factor(-235)
-
-    @property
-    def JACKET_WIDTH(self):
-        return self.apply_factor(375)
-
-    @property
-    def MR_RT_RIGHT_FROM_HMID(self):
-        return self.apply_factor(-300)
-
-    @property
-    def MR_RT_WIDTH(self):
-        return self.apply_factor(275)
-
-    @property
-    def MR_RT_HEIGHT(self):
-        return self.apply_factor(75)
-
-    @property
-    def TITLE_BOTTOM_FROM_VMID(self):
-        return self.apply_factor(-265)
-
-    @property
-    def TITLE_FONT_PX(self):
-        return self.apply_factor(40)
-
-    @property
-    def TITLE_WIDTH_RIGHT(self):
-        return self.apply_factor(275)
-
-
-class SizesV2(Sizes):
-    def __init__(self, factor: float):
-        self.factor = factor
-
-    def apply_factor(self, num):
-        return apply_factor(num, self.factor)
-
-    @property
-    def TOP_BAR_HEIGHT(self):
-        return self.apply_factor(50)
-
-    @property
-    def SCORE_PANEL(self) -> Tuple[int, int]:
-        return tuple(self.apply_factor(num) for num in [447, 233])
-
-    @property
-    def PFL_TOP_FROM_VMID(self):
-        return self.apply_factor(142)
-
-    @property
-    def PFL_LEFT_FROM_HMID(self):
-        return self.apply_factor(10)
-
-    @property
-    def PFL_WIDTH(self):
-        return self.apply_factor(60)
-
-    @property
-    def PFL_FONT_PX(self):
-        return self.apply_factor(16)
-
-    @property
-    def PURE_FAR_GAP(self):
-        return self.apply_factor(20)
-
-    @property
-    def FAR_LOST_GAP(self):
-        return self.apply_factor(23)
-
-    @property
-    def SCORE_BOTTOM_FROM_VMID(self):
-        return self.apply_factor(-50)
-
-    @property
-    def SCORE_FONT_PX(self):
-        return self.apply_factor(45)
-
-    @property
-    def SCORE_WIDTH(self):
-        return self.apply_factor(280)
-
-    @property
-    def JACKET_RIGHT_FROM_HOR_MID(self):
-        return self.apply_factor(-235)
-
-    @property
-    def JACKET_WIDTH(self):
-        return self.apply_factor(375)
-
-    @property
-    def MR_RT_RIGHT_FROM_HMID(self):
-        return self.apply_factor(-330)
-
-    @property
-    def MR_RT_WIDTH(self):
-        return self.apply_factor(330)
-
-    @property
-    def MR_RT_HEIGHT(self):
-        return self.apply_factor(75)
-
-    @property
-    def TITLE_BOTTOM_FROM_VMID(self):
-        return self.apply_factor(-265)
-
-    @property
-    def TITLE_FONT_PX(self):
-        return self.apply_factor(40)
-
-    @property
-    def TITLE_WIDTH_RIGHT(self):
-        return self.apply_factor(275)

src/arcaea_offline_ocr/extractor/__init__.py

@@ -1,2 +0,0 @@
-from .common import Extractor
-from .sizes import *

src/arcaea_offline_ocr/extractor/sizes/auto/__init__.py

@@ -1,3 +0,0 @@
-from .common import AutoSizes
-from .t1 import AutoSizesT1
-from .t2 import AutoSizesT2

src/arcaea_offline_ocr/mask.py

@@ -1,119 +0,0 @@
-import cv2
-import numpy as np
-
-from .types import Mat
-
-__all__ = [
-    "GRAY_MIN_HSV",
-    "GRAY_MAX_HSV",
-    "WHITE_MIN_HSV",
-    "WHITE_MAX_HSV",
-    "PFL_WHITE_MIN_HSV",
-    "PFL_WHITE_MAX_HSV",
-    "PST_MIN_HSV",
-    "PST_MAX_HSV",
-    "PRS_MIN_HSV",
-    "PRS_MAX_HSV",
-    "FTR_MIN_HSV",
-    "FTR_MAX_HSV",
-    "BYD_MIN_HSV",
-    "BYD_MAX_HSV",
-    "MAX_RECALL_PURPLE_MIN_HSV",
-    "MAX_RECALL_PURPLE_MAX_HSV",
-    "mask_gray",
-    "mask_white",
-    "mask_pfl_white",
-    "mask_pst",
-    "mask_prs",
-    "mask_ftr",
-    "mask_byd",
-    "mask_rating_class",
-    "mask_max_recall_purple",
-]
-
-GRAY_MIN_HSV = np.array([0, 0, 70], np.uint8)
-GRAY_MAX_HSV = np.array([0, 0, 200], np.uint8)
-
-GRAY_MIN_BGR = np.array([50] * 3, np.uint8)
-GRAY_MAX_BGR = np.array([160] * 3, np.uint8)
-
-WHITE_MIN_HSV = np.array([0, 0, 240], np.uint8)
-WHITE_MAX_HSV = np.array([179, 10, 255], np.uint8)
-
-PFL_WHITE_MIN_HSV = np.array([0, 0, 248], np.uint8)
-PFL_WHITE_MAX_HSV = np.array([179, 10, 255], np.uint8)
-
-PST_MIN_HSV = np.array([100, 50, 80], np.uint8)
-PST_MAX_HSV = np.array([100, 255, 255], np.uint8)
-
-PRS_MIN_HSV = np.array([43, 40, 75], np.uint8)
-PRS_MAX_HSV = np.array([50, 155, 190], np.uint8)
-
-FTR_MIN_HSV = np.array([149, 30, 0], np.uint8)
-FTR_MAX_HSV = np.array([155, 181, 150], np.uint8)
-
-BYD_MIN_HSV = np.array([170, 50, 50], np.uint8)
-BYD_MAX_HSV = np.array([179, 210, 198], np.uint8)
-
-MAX_RECALL_PURPLE_MIN_HSV = np.array([125, 0, 0], np.uint8)
-MAX_RECALL_PURPLE_MAX_HSV = np.array([130, 100, 150], np.uint8)
-
-
-def mask_gray(__img_bgr: Mat):
-    # bgr_value_equal_mask = all(__img_bgr[:, 1:] == __img_bgr[:, :-1], axis=1)
-    bgr_value_equal_mask = np.max(__img_bgr, axis=2) - np.min(__img_bgr, axis=2) <= 5
-    img_bgr = __img_bgr.copy()
-    img_bgr[~bgr_value_equal_mask] = np.array([0, 0, 0], __img_bgr.dtype)
-    img_bgr = cv2.erode(img_bgr, cv2.getStructuringElement(cv2.MORPH_RECT, (2, 2)))
-    img_bgr = cv2.dilate(img_bgr, cv2.getStructuringElement(cv2.MORPH_RECT, (1, 1)))
-    return cv2.inRange(img_bgr, GRAY_MIN_BGR, GRAY_MAX_BGR)
-
-
-def mask_white(img_hsv: Mat):
-    mask = cv2.inRange(img_hsv, WHITE_MIN_HSV, WHITE_MAX_HSV)
-    mask = cv2.dilate(mask, cv2.getStructuringElement(cv2.MORPH_RECT, (2, 2)))
-    return mask
-
-
-def mask_pfl_white(img_hsv: Mat):
-    mask = cv2.inRange(img_hsv, PFL_WHITE_MIN_HSV, PFL_WHITE_MAX_HSV)
-    mask = cv2.dilate(mask, cv2.getStructuringElement(cv2.MORPH_RECT, (2, 2)))
-    return mask
-
-
-def mask_pst(img_hsv: Mat):
-    mask = cv2.inRange(img_hsv, PST_MIN_HSV, PST_MAX_HSV)
-    mask = cv2.dilate(mask, (1, 1))
-    return mask
-
-
-def mask_prs(img_hsv: Mat):
-    mask = cv2.inRange(img_hsv, PRS_MIN_HSV, PRS_MAX_HSV)
-    mask = cv2.dilate(mask, (1, 1))
-    return mask
-
-
-def mask_ftr(img_hsv: Mat):
-    mask = cv2.inRange(img_hsv, FTR_MIN_HSV, FTR_MAX_HSV)
-    mask = cv2.dilate(mask, (1, 1))
-    return mask
-
-
-def mask_byd(img_hsv: Mat):
-    mask = cv2.inRange(img_hsv, BYD_MIN_HSV, BYD_MAX_HSV)
-    mask = cv2.dilate(mask, (2, 2))
-    return mask
-
-
-def mask_rating_class(img_hsv: Mat):
-    pst = mask_pst(img_hsv)
-    prs = mask_prs(img_hsv)
-    ftr = mask_ftr(img_hsv)
-    byd = mask_byd(img_hsv)
-    return cv2.bitwise_or(byd, cv2.bitwise_or(ftr, cv2.bitwise_or(pst, prs)))
-
-
-def mask_max_recall_purple(img_hsv: Mat):
-    mask = cv2.inRange(img_hsv, MAX_RECALL_PURPLE_MIN_HSV, MAX_RECALL_PURPLE_MAX_HSV)
-    mask = cv2.dilate(mask, (2, 2))
-    return mask

src/arcaea_offline_ocr/masker/common.py

@@ -1,55 +0,0 @@
-import cv2
-
-
-class Masker:
-    @staticmethod
-    def pure(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def far(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def lost(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def score(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def rating_class_pst(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def rating_class_prs(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def rating_class_ftr(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def rating_class_byd(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def max_recall(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def clear_status_track_lost(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def clear_status_track_complete(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def clear_status_full_recall(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()
-
-    @staticmethod
-    def clear_status_pure_memory(roi_bgr: cv2.Mat) -> cv2.Mat:
-        raise NotImplementedError()

src/arcaea_offline_ocr/ocr.py

@@ -7,7 +7,6 @@ import numpy as np
 from numpy.linalg import norm
 
 from .crop import crop_xywh
-from .mask import mask_byd, mask_ftr, mask_prs, mask_pst
 from .types import Mat, cv2_ml_KNearest
 
 __all__ = [
@@ -199,13 +198,3 @@ def ocr_digits_by_contour_knn(
 ) -> int:
     samples = ocr_digits_by_contour_get_samples(__roi_gray, size)
     return ocr_digit_samples_knn(samples, knn_model, k)
-
-
-def ocr_rating_class(roi_hsv: Mat):
-    mask_results = [
-        mask_pst(roi_hsv),
-        mask_prs(roi_hsv),
-        mask_ftr(roi_hsv),
-        mask_byd(roi_hsv),
-    ]
-    return max(enumerate(mask_results), key=lambda e: np.count_nonzero(e[1]))[0]

src/arcaea_offline_ocr/sift_db.py

@@ -1,110 +0,0 @@
-import io
-import sqlite3
-from gzip import GzipFile
-from typing import Tuple
-
-import cv2
-import numpy as np
-
-from .types import Mat
-
-
-class SIFTDatabase:
-    def __init__(self, db_path: str, load: bool = True):
-        self.__db_path = db_path
-        self.__tags = []
-        self.__descriptors = []
-        self.__size = None
-
-        self.__sift = cv2.SIFT_create()
-        self.__bf_matcher = cv2.BFMatcher()
-
-        if load:
-            self.load_db()
-
-    @property
-    def db_path(self):
-        return self.__db_path
-
-    @db_path.setter
-    def db_path(self, value):
-        self.__db_path = value
-
-    @property
-    def tags(self):
-        return self.__tags
-
-    @property
-    def descriptors(self):
-        return self.__descriptors
-
-    @property
-    def size(self):
-        return self.__size
-
-    @size.setter
-    def size(self, value: Tuple[int, int]):
-        self.__size = value
-
-    @property
-    def sift(self):
-        return self.__sift
-
-    @property
-    def bf_matcher(self):
-        return self.__bf_matcher
-
-    def load_db(self):
-        conn = sqlite3.connect(self.db_path)
-        with conn:
-            cursor = conn.cursor()
-
-            size_str = cursor.execute(
-                "SELECT value FROM properties WHERE id = 'size'"
-            ).fetchone()[0]
-            sizr_str_arr = size_str.split(", ")
-            self.size = tuple(int(s) for s in sizr_str_arr)
-            tag__descriptors_bytes = cursor.execute(
-                "SELECT tag, descriptors FROM sift"
-            ).fetchall()
-
-            gzipped = int(
-                cursor.execute(
-                    "SELECT value FROM properties WHERE id = 'gzip'"
-                ).fetchone()[0]
-            )
-            for tag, descriptor_bytes in tag__descriptors_bytes:
-                buffer = io.BytesIO(descriptor_bytes)
-                self.tags.append(tag)
-                if gzipped == 0:
-                    self.descriptors.append(np.load(buffer))
-                else:
-                    gzipped_buffer = GzipFile(None, "rb", fileobj=buffer)
-                    self.descriptors.append(np.load(gzipped_buffer))
-
-    def lookup_img(
-        self,
-        __img: Mat,
-        *,
-        sift=None,
-        bf=None,
-    ) -> Tuple[str, float]:
-        sift = sift or self.sift
-        bf = bf or self.bf_matcher
-
-        img = __img.copy()
-        if self.size is not None:
-            img = cv2.resize(img, self.size)
-        _, descriptors = sift.detectAndCompute(img, None)
-
-        good_results = []
-        for des in self.descriptors:
-            matches = bf.knnMatch(descriptors, des, k=2)
-            good = sum(m.distance < 0.75 * n.distance for m, n in matches)
-            good_results.append(good)
-        best_match_index = max(enumerate(good_results), key=lambda i: i[1])[0]
-
-        return (
-            self.tags[best_match_index],
-            good_results[best_match_index] / len(descriptors),
-        )