Designing Scrollsaw Patterns with Paint Shop ProX Part 4 Introduction to Conversions

---

In this tutorial we will take our first steps into the conversion process. This is the process that renders a photo or drawing to 2 colors as well as optimizes the print for the best contrast between the 2 colors. After the conversion process is completed, we then revert back to the principals learned in Part 3 to complete the pattern.

This conversion is where our computer software is most important. While the computer does the actual process, it can only do what you command it do to. Therefore, your input is what determines the ultimate results.

Consider that two people may start with the same identical photo, using the same software on the same computer but end up with two different patterns. The computer didn't design the pattern. The software didn't design the pattern. The operator that supplied the input and executed the commands designed the pattern. There is no "magic bullet" when it comes to creating patterns. Still, people are looking for that elusive program that will somehow perform this magic.

We will also be talking more about setups in this lesson. Taking the time to learn these setups will simplify your work in the future.

---

Start Paint Shop Pro, click the "open" icon on the PA Bar, navigate to the location where you have the BullTerrier.jpg saved, select that image and then click "Open". This will be one of the images which you downloaded before the first lesson. I prefer to maximize the window so that only the image is shown in the workspace. The size of the workspace can also be increase a little by turning off the rulers. (View>Rulers> uncheck)

Now go back to the PA Bar and click the "File Save As" icon, which should be the last icon on the right end of the bar.

In the "Save As" dialog window: Select the location where you want to save the patterns that you are working on. I call my folder "In Progress". File name can remain "BullTerrier" if you like, but most important is the "File Type" that is selected in the bottom drop down list. Here you want to choose "Paint Shop Pro PspImage". This file format creates a considerably larger file size than a jpg or gif format, but by saving as a pspimage all attributes are retained until the pattern is finished and it will then be saved in a different format with a smaller file size requirement. Once the final pattern is saved to another format the pspimage file may be deleted to recover the disk space.

If you do not change the file format of the file you open and you want to preserve the original image, you must use the Save As command, rather than the Save command, to avoid overwriting the original file.

Below are screenshots of what the screen will look like after the "Save As" command; ready to go to work. (Top PSPX, Bottom PSP8)

Before we proceed let's review these screenshots:

• At the very top of the screen is the "Title Bar" which shows the name of the opened file, "BullTerrier", the file extension ".pspimage" and the current level of magnification. Finally, the active layer is indicated in parenthesis
• The next row down is the "Main Menu". This is the standard Windows menu that we are all familiar with.
• Under the "Main Menu" is our workhorse, the "PA Bar"
• And under the "PA Bar" is the "Tool Options" palette which is displaying the options available for the last tool selected from
• the "Tools Toolbar", the toolbar that is docked vertically on the left side of the screen.
• To the right of the screen is the "Materials Palette"
• and the "Layers Palette". There is currently only one layer which is the picture showing in the
• "Workspace", the large area in the center of the screen.
• One final item is shown. Across the bottom of the screen is the "Status Bar". We have not talked much about the status bar, but it is also a handy little tool.

  At the far left side of the status toolbar is displayed the name of the selected tool. When the tool is in the workspace, the status bar gives pointers on how to use the selected tool. If the icons on the "toolbar" are moused over, the status bar will described the tool and the function that it will perform.

  Also shown on the status bar are the X,Y coordinates of the pointer anytime it is over an image and the RGB values for a color that is moused over (PSPX). Finally the size of the open image is shown as well as the color depth.

The above arrangement of palettes and toolbars gives us an efficient layout while not over crowding the screen. There are numerous other available toolbars and palettes but they would not serve any purpose for our work and would only confuse the screen if they were opened.

There are three functions we need to perform before we start the conversion process on any picture, photo or drawing.

• First the color depth needs to be increased to 16 million colors. Go to Image>increase color depth> RGB - 16 bits/channel. (PSP8/9 - 16 million colors (24-bit) If the option is not available the image is already at that level. For more information on Color Depth click here.
• Second we want to duplicate the image. Notice that the "Layers Palette" is showing only one layer. This layer is labeled "Background" and that is referring to the image showing in the workspace. Now go to the "PA Bar" and click on the "Duplicate Layer" icon. Just mouse over the icons until you find the right one. (If you do not get the pop up text tags when mousing over the icons click here.

  After clicking the duplicate layer icon, the "Layer Palette" will indicate that it now contains two layers: Background and Copy of Background. < PSPX - Left PSP8 - Right > More importantly, note that one of the layers is highlighted in "Blue". This highlighting indicates that that layer is the "active" layer. Any changes that are made to the image in the workspace will be made to the "active" layer. To make another layer active, highlight it by clicking on what appears to be two sheets of paper to the left of the layer thumbnail or name. Go ahead and try it, but change the "copy of background" layer back to being the active layer before continuing.

  We will be talking much more about "Layers" and the "Layers Palette" in up coming tutorials but for now it is only important to remember what was just mentioned.

• The function that should be preformed to a new image at this stage is resizing. This subject is somewhat complicated, so we will address resizing in depth in a later study. For the time being, we will just keep the image at its current size.

The image currently on our screen appears to be a dog, but for our immediate purposes it is a "Guinea Pig". lol We will develop a pattern from the image before the end of this lesson but first we will use the image for further exploration. Maybe boring; but it is need to know information none the less.

Turning our attention now to our Guinea Pig Dog image, we see what appears to be more or less a black and white picture, and black and white (or light and dark) are the only two options we have for creating a scrollsaw pattern. One color cuts out, the other stays. One represents the finished piece, the other represents waste. However, our current image is not simply two colors....yet.

Go to Image>Count Image Colors. LOL, not exactly 2 colors. So our image is black and white plus 215 shades of gray. We know that we can make a pattern, and cut out the black and leave the white but what do we do with all those "grays"? This is the purpose for the conversion process. Using the process, we have to make the decision as to which shade of gray will be rendered as white and which shade will be rendered as black. We only have those two options.

So the obvious question here is: What is the process?

There is no simple answer. We will be using, even creating, different processes as we continue our study. The process that we use largely depends upon the image that is to be converted, as there are many variables. For this assignment we will be using the "Threshold" command as the process for converting the image to black and white. While "Threshold" adjustment does a good job, often more control is needed to develop a really good pattern.

The graphic to the right is a portion of our guinea pig image that has been enlarged to 1500%. It is being shown here to illustrate the function of the 'threshold" adjustment. We will be setting a value for pixel lightness in the Threshold control to determine which pixels change to white (those above the threshold) and which pixels change to black (those below the threshold). Values range from 1 to 255. Lower values produce more white pixels; higher values produce more black pixels. Someone may mention here that the image can be converted to black and white or 1bit color by going to Image>Decrease Color Depth> 2 Colors. While this is true, that method leaves us no control over the image and will not necessarily render the best base for our pattern. Threshold adjust does give us that control. [Go ahead and try converting the image to 2 colors if you like. Just click "Undo" to return the image to its current state. For this image, the results are not too bad, you will see why later in this discussion, however, I would not depend upon it being that accurate routinely.]

The graphic below shows our image at default, low and high Threshold settings. While the default setting is very close to an optimum level for this image, the next photo or drawing that you convert may require a setting at or near one of the other extremes.

Okay, that's enough theory. Now the real thing, but let's review the prep work before moving on. This needs to be done with every new pattern started.

• Open the image
• Save as: name.pspimage
• Increase Color Depth to 16 million colors
• Duplicate Image
• [In another tutorial we will discuss resizing, which is another preliminary task that would be done at this point. However, it will be a fairly detailed discussion so we will wait until later to address the subject.]
• Click the "Save" icon on the PA Bar. (Not "Save As")

Assuming that the preparation tasks listed above have been executed, click on the "Threshold" icon on the "PA Bar". The "Threshold" dialog window will open but it may be in one of several different formats. We need to adjust the window so that it looks like this:

• Select all three buttons: "Show/Hide Previews", "Auto Proof" and "Proof". When buttons are selected they appear to be pressed in.
• Click the "Maximize" icon on the Threshold window's title bar.
• At the "Presets" drop down list and select "Default"
• Nearer the bottom of the window click on the "Fit to Window" icon. This allows you to view the entire image in both the "Before" and "After" sub-windows. (Note: PSPXI has only one button - "Preview" in the upper left corner of the dialog window which preforms the above tasks.)
• The Threshold adjustments are made by increasing or decreasing the number values. The numbers can be manipulated by using the up or down arrows to the right of the set value or by using the slider that will appear if you click the down arrow with the line underneath it.

Remember this process for setting up the Dialog window because we will come across other functions that use similar windows as we continue our study. So in the future when the tutorial calls for Maximizing a Dialog window, that will mean adjusting the window as described above.

Now with the Threshold window maximized, set the number values to ????. lol Once again we are reminded that there are no magic bullets and there ain't no Santa Claus.

Of course the point is, not only will the setting be different for each picture you convert but the setting is subject to each individual’s interpretation. But to help us reach a decision as to what the settings should be we do have some criteria to use as a guide.

1. We want a pattern that has good detail. This usually (not always) means more black as we will see that details are lost as we increase the white portions of the pattern
2. Conversely, when we increase the black generally the white areas become more isolated and therefore will be more difficult to tie into other white areas.
3. Black areas add depth. Too little black often leads to an outlined look.
4. White is needed to give the cutting strength. If we eliminate too much white and then depend on long, thin bridges to tie it all together we end up with a weak cutting.

With these guidelines in mind we proceed to "play around" with various settings. Nothing is final until you click "OK" and even that can be undone.

This graphic shows three possibilities:

The picture in the center is at the default settings of "128". This isn't too far off from what we are looking for. The picture on the left is set at "110". At this setting too much white has been brought in which has caused the loss of definition across the back, head, and the right foreleg. At the same time it has closed up and in some cases tied in some of the isolated white areas. But still, a higher setting would be better.

The picture on the right is set at "146", slightly higher than default. I like this setting because it has added some depth to the head and right fore body but still has not caused excessive separation of the white areas. The white areas affected the most are on the front legs and that can be tied in easily. The most difficult areas to work with at this setting will be the grass.

The above observations are only my interpretation, and you may see it completely different. That's ok. As long as an interesting and cutable pattern is produced, there isn't a right or wrong way.

So go ahead, study the various settings and make your choice then click "OK". Click "Save".

That was the conversion process. As I mentioned earlier, the conversion process is the easy part, what you do with the paint brush and paint now is what determines the quality of the finished pattern. After you become familiar with the procedure, you will find that you will be up to this point in your design in 5 or 10 minutes from the time you first open the image. Most of the remainder of the time required to produce a pattern will be using the paint brush.

There are some refinements that we can make while using the Threshold adjustment that will make for easier and possibly better patterns, but we will wait until the next tutorial to discuss these options.

However, I would like to show you a little trick before we go further.

If you will scroll back up the page to the last graphic, the one on the right is the settings that I chose. If your settings were near the same, the definition of the top of the dog’s tail is either weak or completely lacking. We need to use our paint brush and add some additional outline here. The problem is: where exactly do I draw in this tail, detail? (pun not intended lol) Here is the trick.

Looking at the layers palette, make sure that the top layer (Copy of Background) is active. That is, it has the blue highlight. If not, click on the paper icon to make it the active layer. Now locate and click on the slider as indicated on the graphics to the right. While watching the image in your workspace, slowly drag the slider to the left. As you do, you will notice that the original image (bottom layer), will begin to "bleed" thru to the top image. Stop the slider at about "50". Now pick up your paint brush, set the color to black and with the appropriate size brush stroke trace the missing outline of the tail. When satisfied with the results, return the slider back to 100. This is a powerful little trick that you will want to remember as you will be using it often.

Now using the paint brush tool with black and white paint the pattern is completed exactly the same way described in the last tutorial.(Part 3) Although, slightly more difficult, the basic procedure is the same. That being to tie in all the floaters in a visibly pleasing manner while retaining the strength and integrity of the design. Again, this is not a "paint by the number" process but it is a matter of interpretation. I know all too well how discouraging this part can be, but trust me, if you will just practice, practice, practice, it does get easier with time and experience. I often find that it helps to do a "test" fill, as in the graphic to the left, before starting the tying in process. This will give you a general idea where most of the loose parts lay and you can develop a tentative game plan on how to proceed before you begin. Hint: become good buddies with the "Save" icon and click it often.

After you have finished your pattern, do one final fill to test for floaters. If you are then satisfied with the pattern, click the "Save As" icon, and save in your pattern folder in the jpg or gif format. Gif usually results in a smaller file size.

Practice Project: There are more dog images included in the image downloads. Using the same process as outlined above, convert several of those pictures to patterns. The more you do, the more you will learn.

If you are just starting out designing patterns, you may find this lesson difficult, but don't back off. This lesson is a transition point. The next part builds on the skills that you developed in this exercise. So we just keep crossing the bridges one at a time as we get to them, and soon the going will become much easier.

Pattern Academy©2007