Image Search

Quickly download images using the included search providers pre-configured to find the images you need.

Overview

Download NCollector Studio

Free Lite Version or 30-day Free Trial

NCollector Studio includes pre-configured search providers for google images and bing images, optimized to find the images you want. Just specify a search criteria and override the size filtering if needed and you're good to go. You may also create your own search providers as described below.

Features

Creating your own search provider

You may also create your own search providers by creating an xml document with some information of the details of the target website. Here's a complete example of how to create a search provider for Bing Images (see explanation below):

<?xml version="1.0"?>
<searchprovider name="Bing Images" levels="3" minwidth="300" minheight="300" minsize="30" user_agent="NCollector Studio 1.x">
    <queryurls>
    <queryurl value="http://www.bing.com/images/search?q=[SEARCHQUERY]&amp;qpvt=[SEARCHQUERY]&amp;FORM=IGRE#x0y1800"/>
    </queryurls> 
    <extensions>
    <extension value=".jpg"/>
        <extension value=".jpeg"/>
    </extensions>
    <validationrules>
    <validationrule value="view\=detail" />
    </validationrules>
</searchprovider> 

        

First we set up the basics; a name for the search provider, how many levels needs to be crawled from the root url(s), minimum and maximum image size (in Kb) and which user agent to use.

Then we need to find the starting url, and the easiest way to do this is to perform an actual search in Bing Images using your preferred browser. When searching for e.g. "polar bear" you'll see an url looking something like this:

http://www.bing.com/images/search?q=polar+bear&qpvt=polar+bear&FORM=IGRE#x0y1800

This url contains the search criteria which we need to replace with the NCollector Studio replacement string "[SEARCHQUERY]". It also contains an xy parameter which specifies how far we have scrolled in the browser (Bing images does not use paging, but loads content on demand), so at least we need to set y to a decent value like "1800" (px). Lastly we need to escape the & character in order to use it in xml (replace with &amp;).

http://www.bing.com/images/search?q=[SEARCHQUERY]&qpvt=[SEARCHQUERY]&FORM=IGRE#x0y1800

The validationrules section contains a set of regex expressions. The crawler only follows the links on the page if the regular expression matches. In our case all image links we're interested in contains the "view=detail" string, so we use this to restrict the number of links the spider follows.

The final step is to copy the xml to the NCollector SearchProviders folder, and give it an .nsp extension. The search providers are typically placed in:

{My Documents}NCollector\SearchProviders

For another example, that uses paging, look at the included Google Images search provider. The included search providers are located in the NCollector Studio installation folder.