"In the world ye shall have tribulation: but be of good cheer; I have overcome the world." –John 16:33

Month: July 2012

License plate reading: Narrowing the license plate search area

Day 329 of 1000

I had a little time this afternoon to play with the images I generated yesterday.  You can see the previous steps I took to get to this point here.  The first step I took was to find the brakelights based on yesterdays images.  That was not too hard, but I am going to have to spend some time to handle the variability in brakelight sizes, intensity, and shade.  Nevertheless, it is definitely something with which I can work.  Here are the brakelight finds for a couple of the sample images:

Back of car - brake light find Back of pickup - brakelight find

From there, I established a search window within which to look for the plate.  I put the search region at the bottom of the image which is something I will have to adjust, but I should be able to narrow the window down both horizontally and vertically without too much trouble.

Back of car - plate search ROI Back of pickup - plate search ROI

The next step will be to “robustify” the brakelight finder and narrow down the search region for the plate.  It might take me a week or so to get back to this as I have lots of stuff on my day job right now.

License plate reading: Great progress in finding the plate

I spent several hours today working on some of the preprocessing I think might help me find the license plate on the back of a car.  As always with this type of problem it is good to start with a few easy cases to help lay some of the groundwork.  That is exactly what I did.  I got a few of images each of the back of a pickup and the back of a car to use as my sample development images.  Here are a couple of sample images so you can see what I am talking about:

Back of car - original Back of pickup - original

My first thought in looking at the images is that there are a lot more edges in the area of the license plate than in other areas of the image so I ran a Sobel magnitude on the entire scene to see if that idea held water.  The following are the Sobel magnitude images:

Back of car - sobel magnitude Back of pickup - sobel magnitude

I was right about the license plate lettering, the sobel magnitude image shows a high density of edges, but the proble with this is that there are other high edge density areas of the image.  What I needed was a way to narrow down the number of edges so I decided to separate the veritcal edges from the horizontal edges.  The horizontal edge image was pretty worthless, but the vertical edge image diminshed many of the extraneous, non-licencse plate edges while still maintaining high density in the are of license plate.  The following are the vertical and horizontal edge magnitude images for the car:

Back of car - horizontal edge images Back of car - vertical edge images

So we are a lot closer than when we stared.  We could probably do some morphology coupled with connectivity analysis (blobs) and have a pretty good probability of knowing the position of the license plate for these particular cars.  After thinking about it for awhile, I thought I would try one more thing to narrow down the search area for the license plates.  One thing we have going for us is that the license plate for legal cars should always be somewhere between two red tail lights.  So the next step I thought it would be good for us to take is to create an image that maximizes the red channel and suppresses the non-red area of the image.  On these two cars, I got some pretty amazing results:

Back of car - red maximization/non-red damping Back of pickup - red maximization/non-red damping

We got very good results.  Almost everything in the image is dark with the exception of the red tail lights of the vehicles.  The letters on these license plates just happen to be red, so they showed up quite well, too, but not all plates have red lettering.  We will have to see what happens on red cars, too.  Still, license plate lettering has a finite number of colors, so we will be able to use that to our advantage in the future.  I think we are at a point now where I will be able to combine the information from the vertical edge image and the red channel maximized image to start looking for the plate.  I will do some image cleanup (morphology and other filters) along with connectivity analysis or area image statistics to isolate the plate.  I will probably do that next week or whenever I get a chance to get back to this.

License plate reading: thinking about finding the plate

Day 328 of 1000

Today, the kids need to spend most of the day at the Hill Library at NCSU again, so I will have a good block of time to start figuring out how to find the license plate on the back of the car.  I spent some time looking at images of cars and reading about approaches to find the plates.  Like all general machine vision problems, it is harder than it looks.  Sometimes there are no or very few edges that define the outer boundaries of the plate, so it is not generally a good idea to look only for the shape of the plate.  They only thing you can really depend on is that the characters on the plate are of high contrast.  I am going to look for some images that will get me started.  I will probably start with three or four images to get a concept going, then expand my efforts to increasing data sets.  I think I probably should take my camera with me to take some pictures today.  Maybe I will post a few.

Men need to be men

Day 327 of 1000

It is amazing to me that some men do not model masculine behavior to their sons.  Boys should be taught to take responsibility for social and other situations.  It boggles my mind when that does not happen.  It is not only a matter of masculinity, but of good manners to stand up and be counted.

A hot and tiring summer

Day 326 of 1000

We had a few days of heat a week or so back up around 105 degrees.  Then we got a series of thunderstorms that brought the high temperature down to about 85.  That would not be so bad if it were not so muggy.  It is about the opposite here as in Oregon.  In Oregon there are about eight months of drizzle followed by four months of the most beautiful place and weather in the world.  The good months are in the summer.  In North Carolina there are about eight months of good weather (I count the two or three weeks of freezing and snow as good because that is they way I was made), followed by four months of humid inferno.  Maybe what we ought to do is try to figure out a way to live in North Carolina during the Fall, Winter, and Spring, then head out to Oregon for the summer.

It seems kind of crazy that the places that seem to be prospering most in the current economic climate have about the worst weather conditions in the lower 48–Texas, North Dakota, Montana.  What is that all about.  I guess if you are tough, you can handle all conditions climate and economic.

Compressed semesters

Day 325 of 1000

Both Kelly and Christian are taking a compressed semester class this summer.  The classes are normally given over a period of 16 weeks, but during the summer they are given in five weeks.  The classes meet every day of the week for an hour and a half.  Kelly is taking a three hour class in Foundations of Advanced math–a proofs class, so that is here entire class load.  Christian’s class includes a lab, so he spends a LOT of time on campus.  The thing that is kind of surprising to me is that they seem to be assimilating the material very well even though both of them literally need to learn a new language with a new “alphabet” in just five weeks.  They have just about as much homework as they did when they had a full course load spread of 16 weeks.  They are doing OK so far, but the big tests will start next week.

Christian has to get up in his Chemistry lab sessions and explain how he solved problems.  He asked me how to teach.  The only thing I could think to tell him was practice.  He has a stellar Chemistry professor that explains things very, very well.  Kelly is in an even better situation with an emeritus professor who is 75 years old that comes back during the summer to teach on class out of love for the material.  The students in both classes know exactly what to expect and the professors articulate the material very well in class.  What a gift.

Rednecks in a company I hate

Day 324 of 1000

My buddy Brett from my current job and I went out to dinner with four guys from a company I pretty much desipise.  It is a large machine vision company with its headquarters in the Northeast who have practices and attitudes I have found to be very suspect over the years.  They have a bad reputation in the industry and an inflated opinion about the quality of thier products.  After the somewhat restrained business discussions we went out to dinner with these guys and had a great time.  Even though the company is from the Northeast and has the attitude and morality you might expect from a company from that part of the world, all the guys with whom we met were from the Southeast.  One of them even gratuated from the Citadel.  We had a great time.  They were good guys and, I hate to admit it, might have been more redneck than me.  The funny deal is that they were wildly more informed about all things political than anyone I have every met from the area around their world headquarters.  I was very pleasantly surprised and am looking forward to doing business with them in the future.

Follow-up from the Saylor Foundaton

I just received the following clarification from a representative of the Saylor Foundation in response to our previous post about her organization.  We thank Cami Roden for taking the time to explain Saylor’s methods for documenting and certifying the completion of a course.

Hi there,

Thank you so much for including the Saylor Foundation in your blog post! I’m the Community and External Affairs for the Saylor Foundation, and just wanted to quickly clarify your point about ways our students can demonstrate that they took the course. If you click through any of our completed courses, you’ll see that each course starts with a set of learning objectives, which are further broken down across each of the course units. At the end of each course, there is a final exam that tests the student’s knowledge and comprehension of each of the course objectives. Upon passing the final exam, students can download and print a certificate of completion – and those students who are using our ePortfolio system also can show proof of completion via their transcript. I hope this clears up any points of confusion! If you have any questions about Saylor.org, please don’t hesitate to reach out.

My best,

Camie

Rodney Starks’ The Triumph of Christianity

Day 323 of 1000

The Triumph of Christianity: How the Jesus Movement Became the World's Largest ReligionThe last issue of World Magazine had an article on the most recent Rodney Stark book, the Triumph of Christianity.  I have read other books he has authored:  The Rise of Christianity, The Victory of Reason, and God’s Battalions.  He is an amazing author whose non-fiction books on the sociology of Christianity read like novels.  I have not yet read this one.  Stark was a professor at University of Washington for 32 years before he moved on to a position as University Professor and co-director of the Institute for Studies of Religion at Baylor University.

I had bought into a many of the popular myths about Christian culture that permeate our society before I read his books.  Some of the ideas held by society at large with respect to the Crusades, the Dark Ages, and the spread of Chrisitanity are just wrong.  Rodney Stark has done more to debunk these myths than anyone.  Whenever I hear  such myths confidently propounded, I want to just start handing out Stark’s books.  I can hardly wait to read his latest effort.

Do science degrees make less money?

Day 322 of 1000

I read an article today from the Washington Post titled U.S. pushes for more scientists, but the jobs aren’t there.  I know they are not a very credible source–I am kind of embarrassed citing anything from the NYT or the WP, but this one took the cake.  The article is about highly degreed people in scientific fields who do not have jobs.  The problem with the article is that it talks almost exclusively about people working in chemistry, biology, and medical disciplines.  The word math shows up only once in the article while engineer and engineering also only show up once each.  It is especially hilarious that the first person with a “science” degree described in the article got her PhD in neuroscience.

It has been my premise that people who can handle the math (including hard statistics) have plenty of opportunities.  No one capable of doing hard math has lacked for opportunities, even during the persistent and growing under employment and unemployment that is a staple of the Obama recession.  I know this anecdotally from the number of headhunter calls I get asking for help in finding technical programmers.  You will find a list of graphs on the Business Insider website, here, that puts numbers to this phenomenon.

Just another reason to drop your subscription to the Washington Post and get a math, statistics, or engineering degree.

CoffeeVision: Background suppression first pass complete

Day 321 of 1000

Another day at the Hill Library on the NCSU campus.  The kids are studying math and chemistry.  I think I get more done in this enviroment than any other.  Today I was able to align the individual frames from the videos I have been using, perform background suppression and accentuate the things in the image that are in motion.  The result is pretty noisy, but we know things about the items we want to find in the image that will help us isolate them to tell us the information we want.  I do not have any pictures to put up today because I need to do some additional pre and post processing to make them show what we have done.  Still, I am very satisfied and still might have something to show by the end of the weekend.

Great summer classes

Day 320 of 1000

I am really glad Christian and Kelly both chose to take a class this summer.  Both of them needed it to be able to graduate in two years.  Actually they need to do the same thing again next summer, but I think they are getting more benefit out of these first classes than just the learning and the credit.  Kelly’s Foundations of Advanced Math class is taught by an emeritus professor who is well over 70 years old.  He tells the students EXACTLY what to expect in the class including how many hours they should study for each hour they are in lecture, how it is important that they bother their professors to make sure they are “getting” the material, exactly what they need to do to get an A, etc.  He gives them extra-credit points on what appears to be a regular basis.  What a great first professor.

Christian is taking a required Chemistry course.  It requires more time at school because it includes lab and recitation classes.  Yesterday, he got an email about a help session to help go over material that will be on his first test on Monday.  It is even more hours at school, but he is there now.  His professor got great reviews on RateMyProfessor.com and it appears the reviews were justified.  We do not know how this will end, but it has started well and there is definitely a lot of support infrastructure to help the students through the material.  If they take advantage of all the help and maintain focus, there is no reason they should not do well.

An online education – Udacity and the Saylor Foundation

Day 319 of 1000

Update:  See clarification from Saylor Foundation here on how they provide certification of course completion.

I just read an article about a company named Udacity that provides free, college level education on line.  It does not have a lot of content yet, but, if I get the concept, it sounds better than some other on line educational systems.  The idea is this:

  • Anyone can take any class they want any time they want for free.
  • After the course is completed, the student can go to a walk-in Udacity testing center to take a test to show they actual get the material and it was not someone else who took the course for them.

If you are into education solely for the sake of learning stuff, then their are probably better sites around.  Andrew pointed us to the Saylor Foundation website.  It has full blown college classes by great universities, but no way for a student to unequivocally demonstrate they took the course and understand the material.  If the Saylor Foundation added some say to do that, they would be way ahead of Udacity and have a way to make more money.  They same is true for Khan Academy.  Stanford and MIT also have classes on line.  It seems like they would benefit greatly by following this model, too.

I think this is the future.  The whole issue with on line education is that it is difficult to prove you were the one who actually took the class.I think these are tremendous ideas to which there will be additonal innovation both in terms of content delivery and confirmation that the student has actually learned the material and been certified in a manner that is appreciated by industry and even academia.

Kelly trains for future cartooning

Kelly has been working on her portrait skills by sketching her friends. Really, she wants to cartoon, but she has such a heavy school load that she cannot work it in other than about one drawing per night and a few on the weekend.  She figures if she keeps up her drawing skills for the next couple of years she will be ready to dive in again on her cartooning during graduate school.  Kelly started drawing a set of comic strip characters based around a little girl named Betty Blonde since she was nine.  Then she drew and published a daily comic strip for two full years starting when she was fourteen. You can check it out hereBetty Blonde

Coffee vision — reading videos

Day 318 of 1000

Happy 4th of July!  I got up early to work on our CoffeeVision project.  Yes, it has a name.  I picked CoffeeVision because I couldn’t think of anything else.  If it comes to anything, we will probably change the name.  We are still very early in the process.  I got most of this done before the family all went out to McDonalds for lunch.  It has been a great day so far.  Today, I added the ability to import videos to the machine vision image processing buffer a frame at a time.  We need a framework for development and testing.  This is it.  I have added a YouTube video to show and explain what we did.  I need to add just a little more infrastructure before I can start adding code to analyze the images.

An excuse to put up a picture of a friend

Day 317 of 1000

Jonathan WrightKelly drew this picture of our wonderful friend Jon the other day and put it up on Facebook.  I thought it was good enough that I did not just want to put it up there, so I put it here.  Jon writes a blog (a little bit sporadically, but well worth a daily visit to see if there is something new).  He writes very, very well has some great pictures of Chile and Brazil.

Caprese Salad

Day 316 of 1000

Caprese saladLorena’s thing lately is Caprese salad. It consists of fresh tomato slices topped with a thing slice of fresh mozarella cheese topped with a basil leaf all with olive oil and a little bit of salt and pepper. It is VERY good and probably pretty healthy. Saturday night, though, we went over to dinner at my old boss and friend, Igor’s house. For an appetizer, the served something that reminds me of the Caprese salad, but, at the expense of possibly offending Lorena, might have been even better.

They started with a sauteed slice of eggplant, topped with a slice of fresh tomato, a basil leaf, and Parmesan cheese.  It was also had olive oil and some spices that included garlic.  It was awesome.  I know I ate way to much of it, but it was health, right?  Now Lorena has decided that we need to start eating more sauteed eggplant.  All of us are in complete agreement.

The Mexican election is today

Day 315 of 1000

This is the first time in many election cycles that my father-in-law did not work in one of the voting casillas.  This is a very big election with what appears to be no good choices.  My biggest fear in this election is that the people are so feed up with years of corruption, they will vote for anyone other than the two main parties, the PAN and the PRI.  The problem with that is the man they might elect is a good friend of Hugo Chavez and could take Mexico down the path taken by Cuba and Venezuela.  That would be a very sad turn of event.  Lorena will call her mom in a couple of hours to hear how it went.  She will call her dad now to get his take on the whole thing.

Page 2 of 2

Powered by WordPress & Theme by Anders Norén