sui_feng/AzurLaneAutoScript
1
0
Fork 0
mirror of https://gitee.com/sui-feng-cb/AzurLaneAutoScript1 synced 2026-03-19 03:44:06 +08:00
Code Activity

Add: Map detector wrapper

Browse Source 
- Add: Basic function to utils
- Fix: Crop grids with edge when using homography
- Opt: use getPerspectiveTransform instead of findHomography
This commit is contained in:
LmeSzinc
2020-07-10 01:35:25 +08:00
parent 887707dc74
commit 04bb282e8c
4 changed files with 108 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
module/map_detection/homography.py
View File

@@ -1,6 +1,5 @@
import time
import cv2
import numpy as np
from PIL import Image, ImageDraw, ImageOps
@@ -141,22 +140,14 @@ class Homography:
# Generate perspective data
src_pts = np.array(src_pts) - self.config.DETECTING_AREA[:2]
dst_pts = np.array([
[0, 0],
[size[0] * self.config.HOMO_TILE[0], 0],
[0, size[1] * self.config.HOMO_TILE[1]],
[size[0] * self.config.HOMO_TILE[0], size[1] * self.config.HOMO_TILE[1]]
], dtype=float)
dst_pts += src_pts[0]
homo, _ = cv2.findHomography(src_pts, dst_pts)
dst_pts = src_pts[0] + area2corner((0, 0, *np.multiply(size, self.config.HOMO_TILE)))
homo = cv2.getPerspectiveTransform(src_pts.astype(np.float32), dst_pts.astype(np.float32))
# Re-generate to align image to upper-left
area = self.config.DETECTING_AREA
area = np.array([[area[0], area[1]], [area[2], area[1]], [area[0], area[3]], [area[2], area[3]]])
area -= self.config.DETECTING_AREA[:2]
area = area2corner(self.config.DETECTING_AREA) - self.config.DETECTING_AREA[:2]
transformed = perspective_transform(area, data=homo)
transformed -= np.min(transformed, axis=0)
homo, _ = cv2.findHomography(area, transformed)
homo = cv2.getPerspectiveTransform(area.astype(np.float32), transformed.astype(np.float32))
size = np.ceil(np.max(transformed, axis=0)).astype(int)
self.homo_data = homo
@@ -177,9 +168,7 @@ class Homography:
self.image = image
# Image initialization
image = crop(image, self.config.DETECTING_AREA)
r, g, b = cv2.split(image)
image = cv2.add(cv2.multiply(cv2.max(cv2.max(r, g), b), 0.5), cv2.multiply(cv2.min(cv2.min(r, g), b), 0.5))
image = rgb2gray(crop(image, self.config.DETECTING_AREA))
# Perspective transform
image_trans = cv2.warpPerspective(image, self.homo_data, self.homo_size)
@@ -366,10 +355,16 @@ class Homography:
"""
Yields (tuple): ((x, y), [upper-left, upper-right, bottom-left, bottom-right])
"""
area = [
self.left_edge if self.left_edge else 0,
self.lower_edge if self.lower_edge else 0,
self.right_edge if self.right_edge else self.homo_size[0],
self.upper_edge if self.upper_edge else self.homo_size[1]
]
x = np.arange(-25, 25) * self.config.HOMO_TILE[0] + self.homo_loca[0]
x = x[(x > 0) & (x < self.homo_size[0])]
x = x[(x > area[0]) & (x < area[2])]
y = np.arange(-25, 25) * self.config.HOMO_TILE[1] + self.homo_loca[1]
y = y[(y > 0) & (y < self.homo_size[1])]
y = y[(y > area[1]) & (y < area[3])]
shape = (len(x), len(y))
points = np.array(np.meshgrid(x, y)).reshape((2, -1)).T