# Posts Tagged ImageJ

Using ImageJ, I counted the number of virus particles in this image and measured their diameter. It was interesting when I realized I could count different objects just by changing the pixel size requirements.

Tags:

Sea Turtles

Tags:

For my ImageJ project, I counted the number of birds in the following image (Image Source):

I decided to analyze them as masks since ellipses did not make sense for birds. The results gave a count of 140 birds. I believe the count was lower than normal because some birds overlap.

Source:

For my ImageJ project, I counted the number of cows in the following image

(source http://sumitnagi.files.wordpress.com/2011/03/img_0435.jpg)

Strategy:

The image was reduced to 8-bit greyscale, then thresholding was done by using Image > Adjust > Threshold. Finally, the birds were counted by by doing Analyze > Analye Particles. I decided to analyze them as masks since ellipses did not make sense for birds.

Data

Count Total Area Average Size Area Fraction Mean

140 91798.000 655.700 1.3 254.577

Inferences:

The total count given after setting the image to 8-bit and doing a particle count was 140. I believe this count was under the true number because some birds overlap.

Tags:

I unfortunately had to miss a lot of the presentation because of computer problems. But i was able to still analyze an image. I found an image of sheep (very original) and analyzed it. I changed it to 8- bit then made it black and white. Then I inverted the image and counted the sheep. Image J is quite an interesting program.

Tags:

I used Image J to analyze a map of New Jersey, I live in Northern New Jersey so I often drive what seems to me as the entire length of New Jersey which averaging 70mph takes about 3 hours. As a result I assumed that NJ is most likely around 200 miles long. In addition I drive around Morristown a lot and that usually takes about twenty to thirty minutes averaging a speed of 50mph, so I assumed Morristown was about 30 miles long. Using the measuring tools we employed with the tree ring picture I was able to measure the actual length of NJ and Morristown and compare. NJ is 168.916 miles long according to Image J and Morristown is 30.708 miles long. Image J was useful here in allowing me to verify and it was nice to see that my internal deductive measurements were relatively accurate. In conclusion, ImageJ is a great tool for measuring, especially when a scale is unavailable. Based on this assignment I believe Image J is probably a useful tool that is likely used in Satellite imaging and in Satellite maps. In addition it is great for counting as we saw with the sheep especially since it has features that allow it to filter out stray pixels and other marks.

ImageJ pic analysis

Tags:

This entry is associated with using ImageJ.  Using google I found an image of a flock of birds:

I then loaded this image into ImageJ and transformed it into a black and white 8-bit version (trimming out erroneous bits).  After which I analyzed particles which produced the following image:

The above image came along with data.  The relevant data being the count which was 15.  Counting birds within a flock can become relatively more difficult when multiple birds are grouped together.  When information is taken away from the image multiple birds would get grouped into the same pixel set.  There are likely tools available to address this sort of issue within ImageJ.

I decided that I didn’t really much care for the bird one… so following the same procedure I did this again except for something that I hate doing manually.  Counting E. coli colonies.

The first image below is the original and the second is the analyzed image 185 colonies in total:

Tags:

Using ImageJ Analysis, I was able to take a picture of a parking lot and determine how many cars would be able to safely park in this area.  Even though it can be easily determined due to the small amount of spaces, this was a good way for me to make sure I was accurately using the ImageJ software. I estimated that each car was roughly 5ft, meaning that each space was about 6 ft. Then, I measured the total length of each vertical line (which was separated by the yellow horizontal line). The total length of each vertical line measured 12.4ft meaning that 2 cars should be able to fit; one on each side of the yellow. I also measured the “space” each car should fit to be approximately 3.6ft. Taking the total length of the horizontal yellow line (25.4ft), I was able to determine that 7 cars should be able to park safely along the length of the yellow horizontal line. Therefore, a total of 14 cars should be able to park in the area of the picture (half spaces were not counted). I used the software to determine this but was also to visually see that 14 cars should be able to park in the spaces. This allows me to trust the software more which will make me more prone to using it.

This is the link to the picture that was analyzed. Measurements are also shown. Previous ImageJ projects are shown at the top. The image discussed in this post is at the bottom of the word document.

ParkingLotWord

Tags:

This is a flock of birds in flight taken at the Sacramento National Wildlife Refuge.  Using ImageJ, I opened this image, set the image type to 8-bit, and adjusted the threshold to create a black and white image.  Then I analyzed the particles to find out how many birds are in the photograph.  This analysis generated an image where each bird was outlined and counted (the second image of black outlines of the birds).

The data showed that there were 38 birds in this photo.  The actual number is higher, about 45 birds in whole or partial.  The reason for the lower count generated by ImageJ is likely due to overlap of birds.  Therefore, two to three birds may be counted as one bird due to the flattening and blocked color of the analyzed image.

ImageJ is a fantastic tool and would be hugely helpful in counting many small objects in an image. For example, ImageJ will be useful in counting the number of visible colonies on a Petri dish, the number of cells in a slide, etc.

Another practical application of ImageJ is scaling.  One can set a scale in an image (the example we worked through in the workshop is of tree rings), and use ImageJ to measure the length of each tree ring once the scale is set.

Tags: