OCRs image file to plain text with tesseract.

ocr.ps1

# ocr tif/png to txt
# requires tesseract

Param(
    [string]$ext = "tif",
    [string]$indir = ".",
    [string]$outdir = $indir,
    [string]$tesseract = "C:\utils\tesseract\tesseract.exe"
)

if (!(test-path $outdir)) {
    mkdir $outdir
}

$files = get-childitem -path $indir -filter *.$ext

foreach ($file in $files) {
    $o = "$outdir\{0}" -f $file.BaseName
    $arguments = "`"$file`" `"$o`""
    start-process $tesseract $arguments -wait -NoNewWindow
}