# Find borders between books in image

**URL:** <https://discourse.processing.org/t/find-borders-between-books-in-image/2568>\
**Category:** Processing\
**Created:** [August 12, 2018, 10:31am UTC](https://discourse.processing.org/t/find-borders-between-books-in-image/2568 "2018-08-12T10:31:48Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![TimeMen](https://avatars.discourse-cdn.com/v4/letter/t/9de0a6/32.png) [@TimeMen](https://discourse.processing.org/u/TimeMen)\
**Post date:** [August 12, 2018, 10:31am UTC](https://discourse.processing.org/t/find-borders-between-books-in-image/2568/1 "2018-08-12T10:31:48Z")

</div>

Hey there,

I want to realize the following with Processing or Processing.py:

1. Load picture of my bookshelf into the processing application.
2. The application should find borders between each two books.
3. Then each book should become an object in a dictionary/array.
4. In addition, the height and main colour of the object should be saved in the dict/array.
5. At the end, the application should sort the objects (books) by colour our height.

Im not sure about whether something like this is possible to realize with Processing.  
Because I have some more knowledge with Python than with Java, It would be great for me to make that project using Processing.py.  
Which Library would be good and do you have any link or tutorial for that kind of image editing?

Thanks for your time,  
Best wishes,  
TimeMen (Jakob).

---

<div class="post-metadata">

**Author:** ![jb4x](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.processing.org/jb4x/32/789_2.png) [@jb4x](https://discourse.processing.org/u/jb4x)\
**Post date:** [August 12, 2018, 10:57am UTC](https://discourse.processing.org/t/find-borders-between-books-in-image/2568/2 "2018-08-12T10:57:01Z")

</div>

Hi,

You can find most of the libraries here: [https://processing.org/reference/libraries/#video&vision](https://processing.org/reference/libraries/#video&vision)  
You can find some interesting ones in the `Video & Vision` section.

For example, the `Image processing algorithms` is implementing a version of the `Canny's algorithm` that is useful to find point of interest in an image.

You can also try to do it without using any library, it possible 🙂

---

<div class="post-metadata">

**Author:** ![TimeMen](https://avatars.discourse-cdn.com/v4/letter/t/9de0a6/32.png) [@TimeMen](https://discourse.processing.org/u/TimeMen)\
**Post date:** [August 12, 2018, 11:02am UTC](https://discourse.processing.org/t/find-borders-between-books-in-image/2568/3 "2018-08-12T11:02:52Z")

</div>

Hey jb4x,  
Thanks for your quick reply.  
Without any library would be impossible for me and there’s not enough time to become a pro, because of my job…  
But I’m going to look for the libraries and website you mention in your reply.  
Hopely I get enough information for my project.

TimeMen

---

<div class="post-metadata">

**Author:** ![Chrisir](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.processing.org/chrisir/32/45_2.png) [@Chrisir](https://discourse.processing.org/u/Chrisir)\
**Post date:** [August 12, 2018, 11:55am UTC](https://discourse.processing.org/t/find-borders-between-books-in-image/2568/4 "2018-08-12T11:55:21Z")

</div>

Can you do a image?

Is it one shelf or several above each other?

---

<div class="post-metadata">

**Author:** ![TimeMen](https://avatars.discourse-cdn.com/v4/letter/t/9de0a6/32.png) [@TimeMen](https://discourse.processing.org/u/TimeMen)\
**Post date:** [August 12, 2018, 12:00pm UTC](https://discourse.processing.org/t/find-borders-between-books-in-image/2568/5 "2018-08-12T12:00:43Z")

</div>

It’s a bookshelf where around 20 books standing side by side.

---

<div class="post-metadata">

**Author:** ![jb4x](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.processing.org/jb4x/32/789_2.png) [@jb4x](https://discourse.processing.org/u/jb4x)\
**Post date:** [August 12, 2018, 12:01pm UTC](https://discourse.processing.org/t/find-borders-between-books-in-image/2568/6 "2018-08-12T12:01:59Z")

</div>

Wouldn’t it be quicker to highlight the book spot by hand?

---

<div class="post-metadata">

**Author:** ![TimeMen](https://avatars.discourse-cdn.com/v4/letter/t/9de0a6/32.png) [@TimeMen](https://discourse.processing.org/u/TimeMen)\
**Post date:** [August 12, 2018, 12:03pm UTC](https://discourse.processing.org/t/find-borders-between-books-in-image/2568/7 "2018-08-12T12:03:47Z")

</div>

If it would be done by the application it would be the perfect solution, but marking it manually could also be a solution worth to realize.  
Let’s see how I can code this ^^