Composing Complex Images in Layers

Using layers in a map we can create complex images by building up images within layers.

 

images\img_rgba_bronze_eg06.gif

 

This image is composed of eight layers, beginning with our sample bronze image.

 

images\img_rgba_bronze_eg01a.gif

 

We used the bronze image to make selections of the sky, clouds, the monument and the pedestal and to save them in the Selections pane. The illustration above shows the clouds selection, which was built up using Touch - Select and Add to Selection mode. These selections were saved as masks for future use. The selections were used to copy pixels out of the initial bronze image and to paste them as new images.

 

images\img_rgba_bronze_eg01.gif

 

The first layer on top of the base image is an image created by taking an image with just the white clouds, colorizing them and then applying a Gaussian Blur to make a halo. This layer was moved up a few pixels relative to the rest of the image.

 

images\img_rgba_bronze_eg02a.gif

 

The clouds image by itself is transparent except for the clouds.

 

images\img_rgba_bronze_eg02.gif

 

When it is placed in a layer over top of the halo, the clouds cover the halo except where it extends above them and in the gaps between clouds where the legs of the monument are located. These gaps will be covered by layers above.

 

images\img_rgba_bronze_eg03.gif

 

The next layer is the drop shadow for the monument, which has been moved a few pixels to the right and down.

 

images\img_rgba_bronze_eg04a.gif

 

This provides a credible shadow effect when the monument itself is added in a layer above.

 

images\img_rgba_bronze_eg04.gif

 

We now will add images in layers above for drop shadows for the text as well as the pixels for the text.

 

images\img_rgba_bronze_eg05.gif

 

Like all the drop shadows, this shadow was created from the text image using Gaussian Blur.

 

images\img_rgba_bronze_eg06.gif

 

The "Manifold" text line was created by making a screen shot of the Manifold text and pasting it into an image. We could then use SHIFT - Select Touch to select all the red pixels that made up the letters, choose Edit - Select Inverse to select all the pixels except the red letters and then choose Edit - Delete to make everything except the text invisible pixels.

 

images\img_rgba_bronze_eg07a.gif images\img_rgba_bronze_eg07b.gif

 

One nuance of how we created the image is that the drop shadow for the monument covers the upper part of the pedestal, as can be seen in the illustration at left above. To remedy this, we selected the pedestal pixels, copied them and then pasted them as a new image layer at the top of the layer stack. As seen in the illustration at right above, this covers up the drop shadow that we do not want in front of the pedestal.

 

images\img_rgba_bronze_eg08.gif

 

Images composed of many layers using a map are very easy to change. We created the above image by moving the Manifold text and drop shadow image layers below the monument and monument drop shadow layers in the map's layer stack. We then moved both the Manifold text image and the text drop shadow image up relative to the other images.

 

Combining Tiled Images into One Image

 

At times we may wish to mosaic or tile several different images into a single image. Perhaps, for example, we have several aerial photographs of a region that together could form a single image. There are two cases to consider. The first case is where we have georegistered images that can be seen in their correct geographic position within a map. The second case is non-georegistered images that when seen together as layers in a map will appear all on top of each other.

 

Combining images into a single image (georegistered images)

 

1. Assemble all the images in a map, each in its separate layer.

2. Make any adjustments in contrast, brightness or other visual characteristics to provide an even visual appearance between adjacent images.

3. Create a single image from the tiled map stack by using the Tools - Make Image command.

 

Combining images into a single image (non-georegistered images)

 

1. Assemble all the images in a map, each in its separate layer.

2. The images (see Repositioning Layers ) to their desired locations.

3. Make any adjustments in contrast, brightness or other visual characteristics to provide an even visual appearance between adjacent images.

4. Create a single image from the tiled map stack by using the Tools - Make Image command.

 

Manifold is different than programs like PhotoShop, where one has a Merge Layers command. The reason is that all images in Photoshop are the same pixel size so one can merge layers by simply overlapping pixels.

 

In Manifold, images can have all sorts of different pixel sizes. a 400 x 600 pixel image could be georegistered to cover the entire Earth or it could be a 1200 DPI image. For that matter, an 800 x 1000 pixel image could cover the same surface area as the 400 x 600 image. We could see these two images tiled next to each other in a map even though the one image has "smaller" pixels than the other one. To merge them into one image, we have to say what pixel size we want that final image to have. This is stated by implication when we specify the size of the combined image in the Save Image to dialog.

 

Tech Tip

 

Sometimes mosaic images overlap each other but do not match well at their edges due to photographic distortion or other effects. In such cases we can achieve a better visual blending of the images if we convert the "upper" image to an RGBa image and then use pixel transparency effects to "feather" the edge of the image from full opacity to full transparency. This can provide a blending effect that gives a better visual match to the adjacent image than a hard edge. See the RGBa Pixel Transparency and Create a Circular Feathered Image topics for relevant techniques.