Everyone looking for a car database with pictures is invited to discuss how I should do this job!
Few years ago I was thinking to add photos collected from Wikipedia in the book-style Car Models Encyclopedia.DOC but I was never sure if this is the right thing to do! .DOC format proved unpopular.
Update: in 2015 I learned web scraping, allowing me to quickly create new databases by copying data from various websites. The new databases for India, Middle East and Australia, made via scraping, do contain image URL, and if you want to bulk download all image files, copy URLs from my database into Tab Save extension for Chrome.
2013 idea… should I collect car photos from Wikipedia?
There are plenty of images on Wikipedia that anyone can collect himself, but numerous people asked ME for a database of car images. Probably these people are too busy to dig for images on Wikipedia, or they want all images to be cropped at same size.
OK, I want to help you! But what I should do? Do I should collect photos of every car model, resize them at same resolution, and sell them as .RAR archive? Should be linked some way with the Excel database?
I love working with fixed data, but in case of photos there are a lot of variables. On the internet you can find cars images taken from various angles, showing cars of various colors, in different places (driving / parked / showroom), what images I should collect? Then, if the car have multiple body variants, the photo should be the base variant, random variant or one photo of each version?
The Car Models & Bodies Database contains over 3000 car models body versions produced 1945-present, so theoretically I need to collect 3000+ photos?
All these dilemmas should be decided BEFORE starting this megalithic work in wrong way… and get disappointed to not see anyone purchasing it.
During early 2013, 4 people contacted me for photo database, but only one told me something: “I dont think that will work? Unless you have a SQL database with pics linked into it.” them “thanks anyways” and quitted the chat… without telling me anything regarding format of photo itself. I hate such lazy people! How I am supposed to know what is the correct format of photo database? I am not familiarized with SQL.
In May 2013, finally one customer told me what to do: find a car website from where find and save photos (but this isn’t a copyright issue??!! I will rather choose free photos from Wikipedia), one photo for each body style, try to choose same angle and environment in all photos, then rename photos with correct car model name, then crop and resize thumbnails (I am confused about this part).
So, I created this: Car Photos Database SAMPLE
90 MB worth of photos, including all Volkswagen models since 1970s, selected photos to be at least 1024×768, side-front angle for most cars, side-back angle for body derivations. The rest of brands will be done after getting more feedback.
Just 1 week passed and another customer told me to STOP the car images database, saying that the above sample is NOT right, the images collected are useless, and may lead to copyright troubles. He was looking to purchase licensed images, made from fixed angles, cleared background… unfortunately I do not know how I can do that. Do you?
Maybe if few customers will say that is OK to collect images from Wikipedia (although may be still copyright issue). Hope you don’t want me to go on street with photo camera to take myself one image of every distinct car model? Who does have time for that, especially hunting rare cars that are never seen in my country?
2015 idea… using web scraping software to get car photos
In 2015 I learned about data scraping from websites… initially scraping only the text data, and created American and Indian car databases as well as Motorcycle database. In 2016 I figured out that I can scrap image URL too, and because customers asked, I added image URL column for the above 3 databases.
So I am selling database of image URL, and if you want to bulk download all image files, use Tab Save extension for Chrome and copy-paste URLs from my database.
In the same way I can use web scraping software to get URL of all images from any website you want. This will save me from spending large amount of time digging for images on Wikipedia.
In December 2016 a customer asked me to scrap an used cars website, to get image URL beside Make, Model, Year. Took only FEW HOURS and I got over 100.000 car images, all in same resolution. He told me to keep it private and do not publish or resell on website. So I am telling you only the idea. If anyone wants to scrap car images in this way, let me know what website to scrap!
I am glad that I did not wasted a week collecting 3000 photos from Wikipedia since there are better methods available.
Car photo survey