diff --git a/pdfminer/pdffont.py b/pdfminer/pdffont.py
index feb85575..9f24afb0 100644
--- a/pdfminer/pdffont.py
+++ b/pdfminer/pdffont.py
@@ -16,6 +16,7 @@
 from .psparser import PSLiteral
 from .psparser import literal_name
 from .pdftypes import PDFException
+from .pdftypes import PDFStream
 from .pdftypes import resolve1
 from .pdftypes import int_value
 from .pdftypes import num_value
@@ -127,7 +128,7 @@ def do_keyword(self, pos, token):
 
 
 NIBBLES = ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.', 'e', 'e-', None, '-')
-
+IDENTITY_ENCODER = ('Identity-H', 'Identity-V')
 
 ##  CFFFont
 ##  (Format specified in Adobe Technical Note: #5176
@@ -648,18 +649,8 @@ def __init__(self, rsrcmgr, spec, strict=settings.STRICT):
         self.cidsysteminfo = dict_value(spec.get('CIDSystemInfo', {}))
         self.cidcoding = '%s-%s' % (resolve1(self.cidsysteminfo.get('Registry', b'unknown')).decode("latin1"),
                                     resolve1(self.cidsysteminfo.get('Ordering', b'unknown')).decode("latin1"))
-        try:
-            name = literal_name(spec['Encoding'])
-        except KeyError:
-            if strict:
-                raise PDFFontError('Encoding is unspecified')
-            name = 'unknown'
-        try:
-            self.cmap = CMapDB.get_cmap(name)
-        except CMapDB.CMapNotFound as e:
-            if strict:
-                raise PDFFontError(e)
-            self.cmap = CMap()
+        self.cmap = self.get_cmap_from_spec(spec, strict)
+
         try:
             descriptor = dict_value(spec['FontDescriptor'])
         except KeyError:
@@ -706,6 +697,36 @@ def __init__(self, rsrcmgr, spec, strict=settings.STRICT):
         PDFFont.__init__(self, descriptor, widths, default_width=default_width)
         return
 
+    def get_cmap_from_spec(self, spec, strict):
+        """
+        For certain PDFs, Encoding Type isn't mentioned as an attribute of
+        Encoding but as an attribute of CMapName, where CMapName is an
+        attribure of spec['Encoding'].
+        The horizaontal/vertical modes are mentioned with diffrent name
+        such as 'DLIdent-H/V','OneByteIdentityH/V','Identity-H/V'
+        """
+        try:
+            spec_encoding = spec['Encoding']
+            if hasattr(spec_encoding, 'name'):
+                cmap_name = literal_name(spec['Encoding'])
+            else:
+                cmap_name = literal_name(spec_encoding['CMapName'])
+        except KeyError:
+            if strict:
+                raise PDFFontError('Encoding is unspecified')
+            cmap_name = 'unknown'
+        if type(cmap_name) is PDFStream:
+            if 'CMapName' in cmap_name:
+                cmap_name = cmap_name.get('CMapName').name
+            else:
+                if strict:
+                    raise PDFFontError('CMapName unspecified for encoding')
+                cmap_name = 'unknown'
+        if cmap_name in IDENTITY_ENCODER:
+            return CMapDB.get_cmap(cmap_name)
+        else:
+            return CMap()
+
     def __repr__(self):
         return '<PDFCIDFont: basefont=%r, cidcoding=%r>' % (self.basefont, self.cidcoding)
 
diff --git a/tests/test_pdfencoding.py b/tests/test_pdfencoding.py
new file mode 100644
index 00000000..396d12d8
--- /dev/null
+++ b/tests/test_pdfencoding.py
@@ -0,0 +1,77 @@
+#!/usr/bin/env python
+
+# -*- coding: utf-8 -*-
+
+import nose, logging, os
+from pdfminer.cmapdb import IdentityCMap, CMap
+from pdfminer.pdffont import PDFCIDFont
+from pdfminer.pdftypes import PDFStream
+from pdfminer.psparser import PSLiteral
+
+class TestPDFEncoding():
+
+    def test_cmapname_onebyteidentityV(self):
+        stream = PDFStream({'CMapName': PSLiteral('OneByteIdentityV')}, '')
+        spec = {'Encoding': stream}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, CMap)
+
+    def test_cmapname_onebyteidentityH(self):
+        stream = PDFStream({'CMapName': PSLiteral('OneByteIdentityH')}, '')
+        spec = {'Encoding': stream}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, CMap)
+
+    def test_cmapname_V(self):
+        stream = PDFStream({'CMapName': PSLiteral('V')}, '')
+        spec = {'Encoding': stream}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, CMap)
+
+    def test_cmapname_H(self):
+        stream = PDFStream({'CMapName': PSLiteral('H')}, '')
+        spec = {'Encoding': stream}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, CMap)
+
+    def test_encoding_identityH(self):
+        spec = {'Encoding': PSLiteral('Identity-H')}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, IdentityCMap)
+
+    def test_encoding_identityV(self):
+        spec = {'Encoding': PSLiteral('Identity-V')}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, IdentityCMap)
+
+    def test_encoding_identityH_as_PSLiteral_stream(self):
+        stream = PDFStream({'CMapName':PSLiteral('Identity-H')}, '')
+        spec = {'Encoding': stream}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, IdentityCMap)
+
+    def test_encoding_identityV_as_PSLiteral_stream(self):
+        stream = PDFStream({'CMapName':PSLiteral('Identity-V')}, '')
+        spec = {'Encoding': stream}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, IdentityCMap)
+
+    def test_encoding_identityH_as_stream(self):
+        stream = PDFStream({'CMapName':'Identity-H'}, '')
+        spec = {'Encoding': stream}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, IdentityCMap)
+
+    def test_encoding_identityV_as_stream(self):
+        stream = PDFStream({'CMapName':'Identity-V'}, '')
+        spec = {'Encoding': stream}
+        font = PDFCIDFont(None, spec)
+        assert isinstance(font.cmap, IdentityCMap)
+
+    def test_font_without_spec(self):
+        font = PDFCIDFont(None, {})
+        assert isinstance(font.cmap, CMap)
+
+
+if __name__ == '__main__':
+    nose.runmodule()