使用该自动脚本之前，需要先安装TESSERACT和ImageMagick，安装使用教程请看如何在Centos下识别图片中的文字并存储到txt中。

#!/bin/bash
# Script name: image2text.sh
# Author: lxy.me (http://lxy.me )
# URL: http://lxy.me/centos-remote-picture-identification-text-script.html
# Description: this script will use tesseract and ImageMagick to convert an image to text.
# Script usage: ./image2text.sh <URL> <output_file>

# Variables
tmp_dir="/tmp" # change this if your temporary directory is *not* /tmp

# Error function
function error {

_error_message=$1

echo "Error: $_error_message"
exit 1

}
# Check number of arguments
[ $# -eq 2 ] || error "Script usage: ./image2text.sh <URL> <output_file>"

# Check that tesseract is installed
[ `which tesseract 2> /dev/null` ] || error "Please install tesseract."

# Check that ImageMagick convert is installed
[ `which convert 2> /dev/null` ] || error "Please install ImageMagick."

# Check that wget is installed
[ `which wget 2> /dev/null` ] || error "Please install wget."
URL="$1"
OUTPUT="$2"
TMP_NAME=`mktemp`

echo ""

if [ -f "$OUTPUT.txt" ]; then

echo -n "Warning: File $OUTPUT.txt already exists. Please press enter to continue, or press CTRL+C to quit now."
read pause < /dev/tty
echo ""
fi

echo "Downloading file: $URL"

wget "$URL" -O "$TMP_NAME-download" > /dev/null 2>&1

# Check wget exit status
if [ $? -ne 0 ]; then error "Unable to retrieve file $URL" ; fi

IMG_CHECK=`identify "$TMP_NAME-download" > /dev/null 2>&1`

if [ $? -ne 0 ]; then

error "Unable to identify image type for $URL."
fi
EXT=`identify "$TMP_NAME-download" | awk '{ print $2 }' | tr '[:upper:]' '[:lower:]' 2> /dev/null`

if [ "$EXT" != "tif" ] && [ "$EXT" != "bmp" ] ; then # Image conversion required

echo "Detected image format: $EXT"
echo "Converting image"
convert "$TMP_NAME-download" "$TMP_NAME.tif" > /dev/null 2>&1
tesseract "$TMP_NAME.tif" "$OUTPUT" > /dev/null 2>&1
echo "Cleaning up..."
rm -f "$TMP_NAME" "$TMP_NAME.tif" "$TMP_NAME-download"

else

echo "Detected image format: $EXT"
tesseract "$TMP_NAME.tif" "$OUTPUT" > /dev/null 2>&1
echo "Cleaning up..."
rm -f "$TMP_NAME" "$TMP_NAME-download"

fi

if [ $? -eq 0 ]; then

echo "Conversion of $URL completed successfully!"
echo "Text has been saved to: $OUTPUT.txt"

else

echo "Conversion of $URL failed. "

fi