I don't know why you have duplicates, try this

#!/usr/bin/perl
use strict;
use DBI;

my $dbfile = 'MapTiles.sqlitedb';
my $dbh = DBI->connect("dbi:SQLite:dbname=$dbfile","","");

my $sth = $dbh->table_info('%', '%', '%');
while (my $r = $sth->fetchrow_hashref()) {
  print join " ",$_,$r->{$_},"\n" for keys %$r;
  print "\n";
}
[download]

Which Operating System are you using ?

Ubuntu Yakkity

Output of your second request

 TABLE_SCHEM main 
TABLE_TYPE INDEX 
TABLE_CAT  
sqlite_sql  
REMARKS  
TABLE_NAME edition 

TABLE_TYPE INDEX 
REMARKS  
sqlite_sql  
TABLE_CAT  
TABLE_NAME image 
TABLE_SCHEM main 

TABLE_SCHEM main 
TABLE_TYPE INDEX 
TABLE_CAT  
sqlite_sql CREATE UNIQUE INDEX index_abcd on image (a, b, c, d) 
REMARKS  
TABLE_NAME image 

TABLE_TYPE INDEX 
sqlite_sql CREATE INDEX index_tileset on image (tileset) 
TABLE_NAME image 
REMARKS  
TABLE_CAT  
TABLE_SCHEM main 

REMARKS  
sqlite_sql CREATE INDEX index_retrieved_tileset on image (retrieved, t
+ileset) 
TABLE_NAME image 
TABLE_CAT  
TABLE_TYPE INDEX 
TABLE_SCHEM main 

REMARKS  
sqlite_sql  
TABLE_NAME sqlite_master 
TABLE_CAT  
TABLE_TYPE SYSTEM TABLE 
TABLE_SCHEM main 

TABLE_SCHEM temp 
TABLE_NAME sqlite_temp_master 
sqlite_sql  
REMARKS  
TABLE_CAT  
TABLE_TYPE SYSTEM TABLE 

TABLE_SCHEM main 
TABLE_CAT  
sqlite_sql CREATE TABLE edition (tileset int, edition int, unique (til
+eset)) 
REMARKS  
TABLE_NAME edition 
TABLE_TYPE TABLE 

TABLE_SCHEM main 
TABLE_TYPE TABLE 
sqlite_sql CREATE TABLE image (a int, b int, c int, d int, tileset int
+, retrieved int, current bool, etag text, size int, data blob, UNIQUE
+(a, b, c, d) ON CONFLICT REPLACE) 
REMARKS  
TABLE_NAME image 
TABLE_CAT  

TABLE_SCHEM main 
TABLE_TYPE TABLE 
sqlite_sql CREATE TABLE version (version int, locale text) 
REMARKS  
TABLE_NAME version 
TABLE_CAT
[download]

Ok, try this

#!perl
use strict;
use DBI;

my $database  = 'TestMapTiles.sqlitedb';
my $tablename = 'image';

# open db
my $dbh = open_db($database);
my $imgfolder = 'maptiles-output';
if (! -d $imgfolder){
  mkdir($imgfolder, 0755) or die "$!";
}

# select data
my $sql = 'SELECT * from '.$tablename;
my $sth = $dbh->prepare($sql);
$sth->execute();

# recreate files
# 0..3 a int, b int, c int, d int, 
# 4..6 tileset int, retrieved int, current bool, 
# 7..9 etag text, size int, data blob,
open LOG,'>',$tablename.'.dat'; 
while (my @f = $sth->fetchrow_array){
  print LOG join "\t",@f[0..8],"\n";
  my $pk = join '_',@f[0..3]; # a_b_c_d primary key
  
  my $filename = "$imgfolder/$pk.png";
#  my $filename = "$imgfolder/$x,$y\@$zoom.png";
  print "creating $filename\n";
  open OUT,'>:raw',$filename or die "$filename : $!";
  print OUT $f[9]; # data blob
  close OUT;
}
close LOG;

# open db
sub open_db {
  my $dbfile = shift;
  my $dbh = DBI->connect(         
    "dbi:SQLite:dbname=$dbfile", # dbname = $dbfile was my mistake 
    "", "", { RaiseError => 1 }       
    ) or die $DBI::errstr;
  return $dbh;
}
[download]

I can't map your filename variables $x,$y\@$zoom.png to the fieldnames so I have used the primary key fields a_b_c_d. All the data (except the blob) should be in the log file image.dat. Maybe you can determine which fields are the x,y map co-ordinates and zoom level.

poj