OCRs image file to text with coordinate info in hocr format with tesseract.

hocr.ps1

# ocr tif/png to hocr (html)
# requires tesseract

Param(
    [string]$ext = "tif",
    [string]$indir = ".",
    [string]$outdir = $indir
)

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

$files = ls "$indir\*.*" -include *.$ext

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