finding duplicate images by name and exif data exif parser from https://github.com/ianare/exif-py

find_duplicate_photos.py

def getdate(f): 
    try:                
        with file(f) as x: 
            exif=EXIF.process_file(x, stop_tag="DateTimeOriginal")
            return exif["EXIF DateTimeOriginal"].values
    except: 
        return None

def noext(f): 
    """ remove ALL extensions from filename (os.path.splitext only removes last) """
    return f[:f.index('.')]

newfiles = [ '1/IMGP1234.JPG', .... ]
oldfiles = [ 'archive/2009/ESWC/... ' ... ]

basenew=collections.defaultdict(list)

for x in newfiles: 
    basenew[noext(os.path.basename(x)).lower()].append(x)

baseold=collections.defaultdict(list)

for x in oldfiles: 
    baseold[noext(os.path.basename(x)).lower()].append(x)

match = []
for base,files in basenew.items():
    if base in baseold:
        for f in files:
            try:
                newdate=getdate(f)
                match+=[(f,x) for x in baseold[base] if getdate(x)==newdate]
            except IOError,e: 
                print e