diff --git a/gallery_dl/extractor/nsfwalbum.py b/gallery_dl/extractor/nsfwalbum.py
index 8f1f3f22..9c4d6860 100644
--- a/gallery_dl/extractor/nsfwalbum.py
+++ b/gallery_dl/extractor/nsfwalbum.py
@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Copyright 2019 Mike Fährmann
+# Copyright 2019-2021 Mike Fährmann
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License version 2 as
@@ -44,16 +44,23 @@ class NsfwalbumAlbumExtractor(GalleryExtractor):
         iframe = self.root + "/iframe_image.php?id="
         backend = self.root + "/backend.php"
         for image_id in text.extract_iter(page, 'data-img-id="', '"'):
-            spirit = text.extract(self.request(
-                iframe + image_id).text, 'giraffe.annihilate("', '"')[0]
-            params = {"spirit": self._annihilate(spirit), "photo": image_id}
+            spirit = self._annihilate(text.extract(self.request(
+                iframe + image_id).text, 'giraffe.annihilate("', '"')[0])
+            params = {"spirit": spirit, "photo": image_id}
             data = self.request(backend, params=params).json()
             yield data[0], {
                 "id"    : text.parse_int(image_id),
                 "width" : text.parse_int(data[1]),
                 "height": text.parse_int(data[2]),
+                "_http_validate": self._validate_response,
+                "_fallback": ("{}/imageProxy.php?photoId={}&spirit={}".format(
+                    self.root, image_id, spirit),),
             }
 
+    @staticmethod
+    def _validate_response(response):
+        return not response.request.url.endswith("/no_image.jpg")
+
     @staticmethod
     def _annihilate(value, base=6):
         return "".join(