msinspectra-blog
msinspectra-blog
Almost There
28 posts
A weekly journey in ECEN70
Don't wanna be here? Send us removal request.
msinspectra-blog · 8 years ago
Text
A Cooler Way to Innovate
After weeks of class suspensions and holidays, finally we’ve been able to meet our professor and discuss again the wonders of signal processing. So, for our latest meeting, we continued studying image processing, but, this time it was all about the applications. 
Each of us in the class researched journals  and previous studies related to image processing and how these authors/ researchers used image processing to their advantage. Some of us have already reported what they found out while the remaining (including yours truly) will continue their respective reports on the next meeting.
Our reports  have proven that the use of image processing varies in different fields and you can almost use it in many aspects. In the field of forensic science, image processing is useful to detect blood in a crime scene  as reported by one of my classmates. We can also use image processing for business like how some use it to detect defects in textile.
In my opinion, it’s commendable that image processing takes medical field into a whole new level. In cases like leukemia,  image processing speeds up the process of knowing the white blood cell count of a person. In conclusion, image processing is a great tool in most field of study. It can also  takes innovations in a great way.
0 notes
msinspectra-blog · 8 years ago
Text
University Games
The annual university games have come again which means a week of fun and rest. Finally, we've been given a chance to unwind and just roam around the campus and do whatever that will make us happy.  In my case, it's eating. However, it was also a tradition that an engineer in the making don’t need sleep. And so, everytime the Ugames are being held, we were always bombarded with lots of thing to do. Namely, we have problem sets and some research designs we need to make for the upcoming week.
And with that being said, goodluck to us and may we have sufficient time to enjoy this year’s annual event.
0 notes
msinspectra-blog · 8 years ago
Text
Get Well Soon
Tuesday again. And as always, it was packed with lots of adventures, project to pass; exams; report, in the same day. But due to an accident, our instructor couldn’t meet us that day. He is stung by a bee and got his face swollen up. i’m hoping for his fast recovery.
0 notes
msinspectra-blog · 8 years ago
Text
Aromatic Coffee Bean
On a bright sunny tuesday, i woke up late and in a panic-mode to at least make it to our dsp lab. It was already my morning routine to check my social media account for announce i might have missed, but since i’m late I deliberately skipped that. That Tuesday was not my lucky day because when i arrived no one, no sir arboleda and no classmate, were there.  good thing I saw one of my classmates and like me, he’s clueless. We decided to go to to the library and then we met some of our classmates. Lately, we found out that sir arboleda announcedn that he will not meet us that day. And the classmate we met in the library are like us, clueless. 
Anyways, since nothing can stop sir arboleda we still have a major activity that needs to be passed next tuesday. I posted image processing before using coffee beans. now, for our next activirty, we need to extract the morphological feature of a coffee bean using image processing. This is not an easy task but I know that it would be fun. so, goodluck to us. :D 
0 notes
msinspectra-blog · 8 years ago
Text
Chances
In the night of Monday, our professor announced that Tuesday will be our research day. Like I told you  in my last post Engr. Arboleda already checked our journals and found some mistakes. Now, here is our chance to set those mistakes right.
0 notes
msinspectra-blog · 8 years ago
Text
Rain or Shine
Due to tropical depression  in the Philippines, some provinces and cities suspended classes. Our province, Cavite, is one of those provinces. And, so, again, might I  add, we weren’t able to meet  our professor. However, rain or shine, our instructor still found a time to discuss lectures. With the help of facebook, He discussed our supposed to be lessons in our fb group. Cool, I know.
Remember my first post?  I mentioned there that we need to convert  old students’ thesis into research journals. Well, it  was due last September 8, 2017. All journals are submitted and Engr. Arboleda already checked a few. He already noticed some mistakes in which he point out in our improvised discussion via facebook.
The first thing he noticed was that some of us used journals from predatory publishers as reference. Now, predatory publishers are questionable, exploitative open-access publishers who uses journals to make a profit. Usually, they charges authors for publication fee and once submitted you cannot withdraw your research journal anymore. These exploits of a publisher are a big no-no for those legitimate publishers. Here are the list of questionable publishers in Beall’s List of Predatory Journal.
http://beallslist.weebly.com/
Next, wrong citations. In TELKOMNIKA, they uses Vancouver format in citations. the right citation should be [1] and not (1) but in using mendeley, the output is (1) and not [1]. He suggested that we can do it manually. Also, there are times that even the references were lacking information. Again, we can edit it manually in mendeley. Another important details we need to remember is that we must not cite in the abstract.
Tumblr media Tumblr media Tumblr media Tumblr media
Mendeley, by the way, is a web program for sharing and discovering research papers. And, you can also manage your journals. If you’re interested, here is their website:
https://www.mendeley.com/
That’s all for now pipz. have a nice weekend ahead.
“Hardwork builds LEGENDS” - Engr. Edwin Arboleda
0 notes
msinspectra-blog · 8 years ago
Text
Filtering/ Extracting the Elements You need in an image
First, you need to input a value for your variable idx to set up your rules, like this:
>> idx=find((400 <=area_values)&(area_values<=800))
As you can see, you set up the rule to only allow only objects with the Areas ranging from 400 to 800.
After this, you can check which numbers/labels you have extracted from your desired image by typing the variable:
idx =
 Columns 1 through 19
    9    13    20    21    22    23    24    25    26    27    28    29    30    31    32    33    36    37    40
 Columns 20 through 24
   43    53    56    62    63
You can then create another variable to finalize your image with your desired elements and objects:
>> h=ismember(f,idx); >> figure,imshow(h),title('KDC: Area between 400 and 800')
This would then give you this output:
Tumblr media
0 notes
msinspectra-blog · 8 years ago
Text
Producing the properties of the elements in your image
To start things up, input this into your command window:
>> g=regionprops(f, 'Area', 'BoundingBox');
By using the command regionprops, the MATLAB will compute the properties of all the elements in your photo’s area, which you can then see by typing the input of the number you want to see, just like this:
>> g(9)
By typing this as your input, it would then pop-up this message, showing the data you want to see, which in this case is the area.
ans =
          Area: 715    BoundingBox: [22.5000 100.5000 28 33]
0 notes
msinspectra-blog · 8 years ago
Text
Using the function vislabels.m
To use the vislabels M-file, let’s input this to our command window:
>> f=bwlabel(e); >> vislabels(f), title('KDC: each object labelled')
This function’s effect is that it will identify all present solid regions in your image and then numbering it from top to bottom, left to right in order. By doing this, you’ll see the different elements that add up to your whole image. Example of this is this output:
Tumblr media
0 notes
msinspectra-blog · 8 years ago
Text
vislabels.m
Here, I’ll show you a function entitled vislabels.m and its code:
function vislabels(L)
background_shade = 200; foreground_shade = 240; I = zeros(size(L), 'uint8'); I(L == 0) = background_shade; I(L ~= 0) = foreground_shade;
% Display the image, fitting it to the size of the figure. imageHandle = imshow(I, 'InitialMagnification', 'fit');
% Get the axes handle containing the image.  Use this handle in the % remaining code instead of relying on gca. axesHandle = ancestor(imageHandle, 'axes');
% Get the extrema points for each labeled object. s = regionprops(L, 'Extrema');
% Superimpose the text label at the left-most top extremum location % for each object.  Turn clipping on so that the text doesn't % display past the edge of the image when zooming. hold(axesHandle, 'on'); for k = 1:numel(s)   e = s(k).Extrema;   text(e(1,1), e(1,2), sprintf('%d', k), ...      'Parent', axesHandle, ...      'Clipping', 'on', ...      'Color', 'b', ...      'FontWeight', 'bold'); end hold(axesHandle, 'off');
0 notes
msinspectra-blog · 8 years ago
Text
Filling holes in your image
To fill the holes in your black and white filtered image, just add this to your command window:
>> e=imfill(d,'holes'); >> figure,imshow(e),title('KDC: Filled Holes Black and White Coffee Beans');
By using imfill and indicating the variable of the image you want to fill, as well as the part of the image you want to be filled which in this case was the holes present in the image, you get this output:
Tumblr media
You can easily identify that this is different from the previous output since this photo’s letter ‘o’ seen in the word Alfonso lost the hole in it which is now also filled with the white filter.
0 notes
msinspectra-blog · 8 years ago
Text
Applying a Black and White Filter
You can create a black and white filtered image by applying this to our previous inputs:
>> d=im2bw(c);
>> figure,imshow(d),title('KDC: Black and White Coffee Beans');
By using this, the command im2bw makes the image stored in the variable you want to modify turn from its color to a black and white type of filter, like this output:
Tumblr media
0 notes
msinspectra-blog · 8 years ago
Text
Creating Complemented Images
To create complemented images, just add this to our previous inputs:
>> c=255-b;
>> figure,imshow(c),title('KDC: Complemented Grayscale Coffee Beans');
By doing so, we get this output:
Tumblr media
0 notes
msinspectra-blog · 8 years ago
Text
Applying Grayscale Filter
To apply a grayscale filter to your image, just input this on your command window:
>> a=imread('coffee.jpg'); >> figure,imshow(a),title ('KDC: Coffee Beans'); >> b=rgb2gray(a); >> figure,imshow(b),title('KDC: Grayscale Filtered Coffee Beans');
As you can see, we used another variable to create another image which is a modified version of the first variable, whereas in our case is the variable ‘a’. By giving the input rgb2gray, the image converts its base color from the “red-green-blue” common base to a grayscale base. Once again, using imshow, we get this output:
Tumblr media
0 notes
msinspectra-blog · 8 years ago
Text
Generating Images
To generate images, just input this in your command window:
>> a=imread('coffee.jpg'); >> figure,imshow(a),title ('KDC: Coffee Beans');
Whereas imread is the command that scans the jpg file that you want to generate. And through this, you can then generate the image by giving the input in the second command line, wherein figure creates a window in your matlab and then followed by imshow of the variable to generate the image you want to appear.
Tumblr media
0 notes
msinspectra-blog · 8 years ago
Text
Image Processing
Hello once again readers! It’s been another week and I’ve got new lessons to share with you. Take note: This week’s topic involves cool outputs.
So this week’s topic, which is already stated in the title, talks about the basics of Image Processing and on how we can utilize this as a powerful tool to analyze different things.
Well then, let’s start.    
0 notes
msinspectra-blog · 8 years ago
Text
Fin.... for now
so that’s it. I have discussed everything I learned about operations on signals. I do hope I explained each topic in a way that a lot of people will understand. Thank you for reading and thank you, as well, for giving me this opportunity to review our lessons.  
0 notes