These are the new features in version 5.0.

• Your choice of settings in the Output tab (the position of the sliders) is preserved when you load a new image to analyze.
• Your choice of color channels in the Analysis tab is preserved when you load a new image to analyze. With these two the user can apply the same settings to a sequence of images if they are similar in nature. So, we get as close as possible to bulk processing without actually creating this complex feature.
• Luminosity is a new color channel that you can choose. It is computed as a combination of the red, green, and blue values: 0.299*R + 0.587*G + 0.114*B. There are four channels now.
• “Display channel” is a new option in the Analysis tab (just like the one in the Output tab). If you have chosen to shrink the image, the shrunken version is shown. This way you can preview all channels and decide which is the best – before committing to time consuming analysis.
• The “Help” menu provides now the links to the help pages of this wiki. The user’s guide and the license are still provided with the program; they are to be found in the “Pixcavator” folder on your hard disk.
• The actual processing time is shown when it’s done, and a beep is produced – but only if processing has taken more than 5 seconds.
• Up to 2000 contours are now shown on the image and their statistics is also displayed. When there are more than 2000 contours, neither is shown.
• A few bugs have been fixed, some remain.

Download here.

This edition is a version of Pixcavator that comes with a preloaded image. Which means that it’s not really a single program but many – one for each image.

It is a single file “exe” program that does not require any installation. As such it can be used as an alternative to screenshots, for demos etc.

This edition has all the features of the standard edition except for image processing tools. This way you can choose different color channels for your analysis, experiment with the sliders, and save your work.

Most of image analysis examples will soon have links to the corresponding files. For a start, try to run these two examples:

• immunohistochemistry,
• melanoma.

In FF: choose “Save”, then “Run”. In IE: just choose “Run”.

We will also be able to create such files as a service to our customers.

The next version of Pixcavator has just been released (download). The update will help the user to create better image analysis reports with Excel.

These reports are based on the data in Pixcavator’s output table that lists all objects in the image as well as their  measurements and locations. The table, combined with the analysis settings, the statistical summary, and the frequencies of the values in the table, can be saved to hard disk in the form of a spreadsheet. This spreadsheet is a complete report of what has happened.

The frequencies allow one to create a histogram for each column, such as the size, to graphically illustrate the distribution of the values. One can also add images, format the text, etc. The end result may look like this:

For more details read Report generation in the wiki.

A note to our current customers: If you purchased Pixcavator within the last 12 months, i.e., after December 13, 2008, you are entitled to a free upgrade. You can download the new version here and then activate it with your current activation number. If your purchase dates before December 13, 2008, you can acquire the new version here. It goes without saying that the older versions of Pixcavator are to remain active indefinitely.

In the late December the price of Pixcavator will be increased to $245.

From other news, our site Computer Vision Primer has reached 268 articles with over 780 illustrations. The fastest growing category is image analysis examples which has reached 56.

Download the new version or first read a bit about the background. 

Besides a few minor improvements, the main novelty is more ways to change the analysis settings. There are six sliders now.

As before we have two sliders for the two main measurements – size and contrast. Now, the confusingly named “Max growth rate” has been remade into “Border contrast”. What does it mean? Both this and the second slider consider the difference in intensity inside the object from that of the surrounding area. But, while the contrast slider considers the maximal difference throughout the object (that why it’s now called Max contrast), the border contrast slider looks at the area just inside the contour. The result is about the same as before: it detects sharp edges, i.e. the contours with the largest difference between what’s just inside and what’s just onside of it.

The next slider works similarly but takes into account the average difference in intensity inside the object from that of the surrounding area. It’s called Average contrast.

The last two sliders simply threshold the image. For example, if you set Intensity, dark to 100, the contours of dark objects will the intensity equal to exactly 100. If you set Intensity, light to 100, the contours of light objects will the intensity equal to exactly 155 (here 155=255-100). It was set this way in order to ensure that contours grow as you move any of the six sliders from left to right. This is a very unsophisticated tool but it’s quite effective when the image lighting is uniform.

The need for introducing these new tools came from a specific, real life image analysis task: Measuring staining in the liver.

Reminder to the current customers: you can download the new version and activate it with your current serial number.

Reminder to the rest: the new price $195 won’t become effective until June 15.

Tracking objects in a video is easy to illustrate with this demo (click the link, a screenshot with explanation is below). The demo comes from an old version of Pixcavator. That “universal” prototype contains the backbone of the current Pixcavator IA, a morphological analysis application (I’ll write about that at some point), and the motion/object tracking application. The program finds objects in each frame of the video but also tells you which ones reappear in the next frame. Both merging and splitting is captured. As a result an object that passes “behind” another is treated as if it is merged. This is not what you expect from object tracking but it actually makes sense: remember there is no “behind” in 2D! To get true tracking one would need to incorporate some image/object matching as well. The program also has other Pixcavator-like features: ignoring objects that are too small or ones that disappear too quicky. It is fairly fast.

The old price, $150, has been the same since August 2007. The new price is $195. It will come into effect in June. Bulk discounts will also be introduced.

This image analysis example came from a biomedical researcher: “I would like to know if the program can calculate relative areas and analyse differences in immunostaining intensity… In images 1 and 2, I need to know if Pixcavator can convert image 1 in 2 and then in image 2 giving a percentage of the are in red in relation to the all field. In image 3, I need 2 things (it is an immunostaining of the liver). I need to know if the program can give me a level of intensity of the immunostaining (that is the intensity of brow in the image) and secondly the fraction of the area that is immunostained (i.e., brown coloured).”

After a few minutes of experimenting with Pixcavator, this is best I can do for now (screenshot #1 below). As you can see some of the red area is not captured while some holes are counted (#2). To find area of those holes I had to analyze the image with different settings (#3). So, the total area of the red is approximately:

10864+3501+1426+2103+1746-1477-1453 = 16710.

Then the percentage is

16710/(853×640) = 3%.

Unsatisfied with the result, I modified the software: I used thresholding as the 2nd slider instead of contrast and redid the analysis. The result was 5.51-1.65=3.86%, much better. 

The third image was trickier because the lighting is uneven. Initially I was unable to capture all brown areas – not simultaneously. Certainly, cropping the image and analyzing parts one by one is an option. (The average color, it is displayed for each contour under “Gray” in the Pixcavator’s output table). That’s something to keep in mind: what is displayed is the average intensity with respect to the color channel that you’ve chosen, in this case “Green”. (The ability to compute the average color with respect to all three channels is still under development.)

Unsatisfied with the result, I came up with another modification: I used average contrast for the 2nd slider and redid the analysis. The result was much better. The average gray (intensity in the green channel) was 132.

The two modifications will appear in version 4.2 due in June.

This version of Pixcavator has only one new feature: the analysis is precomputed. The result is that moving sliders produces immediate effect on the contours and finding the best contours becomes very quick. You can see the effect in this video. The analysis stage is however quite longer than in the previous versions. So, before making this feature permanent I want to get some feedback from the users. That’s why this is just a test release.

Download here. Recent examples here.

The main changes in this version of Pixcavator are the following.

First, an annoying bug in the user interface was fixed. I don’t want to remind everybody what it was but I do certainly apologize. A few minor bugs were fixed too.

Second, a new slider “Border contrast” replaces the old. The idea is that by moving it you can jump to the nest sharp border. For example in the image below, the change in the gray level is very gradual. So, if you move the slider “Size” or “Contrast”, the growth is very slow and the former won’t even notice the sharp edge. With the new slider you get there is just a few abrupt steps: border contrast = 0, 10, and 15, respectively.

To our current customers. You can download the new version of Pixcavator and then activate it with your current serial number. This amounts to free upgrades for a foreseeable future.

BTW, the first digit in the version number refers to the calendar year of the development. This is the fourth since version 1.0 appeared in August ’06 (a prototype/testing program was created in the summer of ’05).

Finally our plans for the coming months:

• 4.1: Introduce a way to fully pre-compute all data (not only the construction of the topology graph but also and its analysis) so that moving sliders will change both the contours and the sliders virtually instantly.
• 4.2: Speed up the core algorithm (construction of the topology graph) significantly. I think its complexity will be linear instead of the current quadratic.
• 4.3: Introduce more data filtering tools (beyond unmark dark and light objects).

I’d be glad to hear your suggestions.

This has been an on-and-off project for almost two years (version 1.0 described here). The purpose is simple: find images similar to a given image. Since it is not even well understood what images are similar, the progress in this area of “image-to-image” search (aka “visual image search”) is very slow–. So, instead, we focus on the goal of finding modified versions of the original. This release is a way to report a limited success we have achieved.The executable PxSearch.exe is accompanied by a small collection of images (download here, 7.2 MB). The system consists of the following modules:

• the collection of images that can be extended; 
• the database containing “signatures” of images, images’ origins, and other data; 
• the image analysis unit (produces the signatures); 
• the matching unit (matches the signatures); 
• user interface (uploads an image, searches for similar images in the collection, displays the matches as a list);

For every image to be added, first the image is converted to grayscale and then shrunk so that the larger dimension is 150. Then several of its secondary versions are created, analyzed, and added to the collection and their data is added to the database, total of 8:

1. original 
2. rotation, 5 degrees 
3. rotation, 45 degrees 
4. Gaussian blur 
5. salt and pepper noise 
6. stretch, 5% 
7. shrink, 5% 
8. crop from all sides, 5%

The entry in the database for each image contains the information about its origin:

• date and time, 
• the filename of the original image, 
• the way the image was produced from the original (shrinking, rotation, etc), 
• the signature of the image.

A signature is a sequence of 126 integers which is the output of image analysis: it is essentially the distribution of sizes of objects found in the image (the data comes form the same source as for Pixcavator).

Suppose the signature of the two images are {An} and {Bn}. Move along these sequences and compute the absolute value of the differences of n-th entries. The result is a distance formula as the “weighted 1-norm metric”:

D = Σ Cn |An – Bn|.

A search is deemed successful if most of the versions of the query image are at the top of the list. This is the case for images that are “good” in the sense that they have clear pattern (based on shapes not color). However, this standard is hard to quantify as it is dependent on the collection. Since the collection I used for testing was small (4500 images), I had to find a way to evaluate the quality of searches that is independent of the size of the collection, as much as possible. So, the quality score for a given image was

(average distance to its 7 versions) / (average distance to all images) * 100.

There are many interesting question to study based on this data and I will report further.

Soon after version 3.2, the next one is here. There are a couple of changes.

First, a feature was added to help with image exploration. As you move the mouse around the analyzed image, the object you hover over is highlighted. The contour is shown blue and, if “Color objects” is chosen, the whole object is colored. The data about it is displayed under the image as before.

Second, the limitations on the kind of objects to be captured have been removed. To avoid dealing with excessive number of objects in the image and the spreadsheet, only objects with “saliency” above a certain threshold were taken into account and displayed. There have been no complaints about that until recently. A manufacturing company needs to analyze an 1000×2000 image which is almost all black with a few light dots here and there (1-3 pixels in size). In the new version even objects this small will be captured. However, the number of objects will be limited to 1000.

The latest version of our image analysis software is out!

Pixcavator was initially designed primarily for counting: cells, objects, and other features. Everything is designed around this task: the clickable contours, the spreadsheet, the averages of the measurements, etc. Now, more and more often we see Pixcavator used for measuring. The reason is simple. Manually measuring the area, or perimeter, of a complex object is close to impossible even if the image is clear and the object is has well defined borders.  A medical example is here and another one here.

Pixcavator captures contours of all objects – light and dark – and displays all their measurements. If, however, there is just one object but with a few holes, it is important to see that this data gives you the area of what’s inside the contour. What you frequently need instead is the area of the object,  which is

the area of what’s inside the contour – the areas of the holes.

The main new feature in version 3.2 is a step in that direction. Pixcavator now displays the two numbers above – the total area of dark and the total area of light – as percentages of the total size of the image (under Review summary, second row). So, to find the area of a dark object with light holes in it, one has to subtract these two numbers. (Caution: You have to make sure however that the holes are in the object not the background).

Download Pixcavator 3.2 here.

The updated interface is the first thing that you notice. All buttons and sliders are arranged in groups accompanied by headers. Text and tooltips were improved throughout.

The RGB channel analysis was completed to include all three channels. Just click a button in the Analysis tab for the color you want.

A new, “Max growth rate”, slider was introduced. Let me explain what it is. As you may remember object in the image are allowed to grow – from one level of gray to the next – up to the extent set by the slider. For example, the object will grow until it’s both larger than say 100 pixels and has contrast above 20. Now, this is a totally different kind of slider. If you choose 10, the object will be allowed to expand – from one level of gray to the next – as long as its size grows by 10% or less. Roughly, the expansion stops once the contour reaches a sharp edge. There will have to be more written about this after some testing. (To reproduce results you obtained with the older versions of Pixcavator just keep this value at 0.)

A new header is Data filtering. There are only two buttons here currently – Unmark dark and Unmark light. For convenience they were redesigned as follows. These are toggle buttons so that you can choose to concentrate on only, say, light objects without having to unmark dark every time you change the settings. There is more to come here.

The analysis summary now includes the mean values and standard deviations of all the main characteristics of objects (marked only).

The way contours are plotted was improved. Now red and green contours never overlap no matter how close they are to each other.

A noticeable speed-up was achieved, in both image analysis and graph analysis part. The memory usage was significantly reduced. There were also numerous minor improvements.

More here.

Pixcavator is a light-weight (336K here) image explorer. Below I list new features and other modifications.

RGB channel-by-channel analysis. It’s an experimental feature, so that you can only use the red or the green for now. This is important for some applications such as microscopy. Different features are sometimes better revealed in different channels. Below: original, analysis in red channel, analysis in green channel.

Analysis summary to include some statistics. The output table contains only the raw data about each object. Of course, if you save the data to Excel, you can get anything from it: average of all columns, histograms, etc. We thought that it would be nice to be able to preview some data: average values of size and contrast. There will be more.

Data displayed based on the location of the mouse. That’s another very convenient feature. You used to have to mark/unmark object in the image and then find the row in the table to see the objects’s measurements. That’s not fun if the table is a hundred rows long. Now you let the mouse hover over the object of your interest and the data from the table is displayed right beneath the image.

Coloring objects. This feature was previewed a couple of weeks ago.

Hiding contours. To see the original image you used to have to go to the Analysis tab. Now you can flick it on and off to see what is hiding under the contours. The marking/unmarking of objects is unaffected.

Some sliders removed. The sliders for roundness and saliency haven’t been used a lot as far as I know. The complexity they add did not seem worthwhile. It does not mean that there will be always just the two sliders. The development of new characteristics for the sliders is under way. They will only be added if they make a significant improvement over what we have now. At least one new slider is coming in the next release.

Shrink slider modified. The shrink slider used to give you the shrink factor in terms of the area of the image. Now if you set itat 2, both of the dimensions will be cut in half while the area (and the processing time) will be cut by 4. This seems simpler. It is also preset to cut the processing time to 10 seconds or less. It seems like 99% of the time the resolution is excessive relative to the features being sought.

The next version of Pixcavator is to be relased in a few weeks. A new feature that I want to preview is Coloring Objects. Once objects are found, you can do anything with them. So, it was easy to implement (the objects are colored randomly). And it’s definitely an amusing feature. It can also be helpful.

This tool can help you confirm that your image segmentation is correct:

A more intricate segmentation:

Coloring combined with background removal:

Something more amusing: recoloring objects and discovering a broken bone:

Just for fun:

For more examples, see our Image Gallery.