kopia lustrzana https://github.com/OpenDroneMap/WebODM
				
				
				
			
		
			
				
	
	
		
			88 wiersze
		
	
	
		
			2.8 KiB
		
	
	
	
		
			Python
		
	
	
			
		
		
	
	
			88 wiersze
		
	
	
		
			2.8 KiB
		
	
	
	
		
			Python
		
	
	
| import glob
 | |
| import os
 | |
| import logging
 | |
| 
 | |
| logger = logging.getLogger('app.logger')
 | |
| 
 | |
| class GCPFile:
 | |
|     def __init__(self, gcp_path):
 | |
|         self.gcp_path = gcp_path
 | |
|         self.entries = []
 | |
|         self.raw_srs = ""
 | |
|         self.read()
 | |
| 
 | |
|     def read(self):
 | |
|         if self.exists():
 | |
|             with open(self.gcp_path, 'r') as f:
 | |
|                 contents = f.read().strip()
 | |
| 
 | |
|             lines = list(map(str.strip, contents.split('\n')))
 | |
|             if lines:
 | |
|                 self.raw_srs = lines[0]  # SRS
 | |
| 
 | |
|                 for line in lines[1:]:
 | |
|                     if line != "" and line[0] != "#":
 | |
|                         parts = line.split()
 | |
|                         if len(parts) >= 6:
 | |
|                             self.entries.append(line)
 | |
|                         else:
 | |
|                             logger.warning("Malformed GCP line: %s" % line)
 | |
|         else:
 | |
|             logger.warning("GCP file %s does not exist" % self.gcp_path)
 | |
| 
 | |
|     def iter_entries(self):
 | |
|         for entry in self.entries:
 | |
|             yield self.parse_entry(entry)
 | |
| 
 | |
|     def parse_entry(self, entry):
 | |
|         if entry:
 | |
|             parts = entry.split()
 | |
|             x, y, z, px, py, filename = parts[:6]
 | |
|             extras = " ".join(parts[6:])
 | |
|             return GCPEntry(float(x), float(y), float(z), float(px), float(py), filename, extras)
 | |
| 
 | |
|     def get_entry(self, n):
 | |
|         if n < self.entries_count():
 | |
|             return self.parse_entry(self.entries[n])
 | |
| 
 | |
|     def entries_count(self):
 | |
|         return len(self.entries)
 | |
| 
 | |
|     def exists(self):
 | |
|         return bool(self.gcp_path and os.path.exists(self.gcp_path))
 | |
| 
 | |
|     def create_resized_copy(self, gcp_file_output, image_ratios):
 | |
|         """
 | |
|         Creates a new resized GCP file from an existing GCP file. If one already exists, it will be removed.
 | |
|         :param gcp_file_output output path of new GCP file
 | |
|         :param image_ratios dictionary with "imagename" --> "resize_ratio" values
 | |
|         :return path to new GCP file
 | |
|         """
 | |
|         output = [self.raw_srs]
 | |
| 
 | |
|         for entry in self.iter_entries():
 | |
|             entry.px *= image_ratios.get(entry.filename.lower(), 1.0)
 | |
|             entry.py *= image_ratios.get(entry.filename.lower(), 1.0)
 | |
|             output.append(str(entry))
 | |
| 
 | |
|         with open(gcp_file_output, 'w') as f:
 | |
|             f.write('\n'.join(output) + '\n')
 | |
| 
 | |
|         return gcp_file_output
 | |
| 
 | |
| 
 | |
| class GCPEntry:
 | |
|     def __init__(self, x, y, z, px, py, filename, extras=""):
 | |
|         self.x = x
 | |
|         self.y = y
 | |
|         self.z = z
 | |
|         self.px = px
 | |
|         self.py = py
 | |
|         self.filename = filename
 | |
|         self.extras = extras
 | |
| 
 | |
|     def __str__(self):
 | |
|         return "{} {} {} {} {} {} {}".format(self.x, self.y, self.z,
 | |
|                                              self.px, self.py,
 | |
|                                              self.filename,
 | |
|                                              self.extras).rstrip() |