The Hybrid Image How to Create It

Learn Photo Editing

Learn Photo Editing

Get Instant Access

Types of Hybrid Images

Mosaics. Why bother with mosaics if wide-field vistas can be imaged using short-focal-length instruments? The answer is that when using a shorter focal length to achieve greater field size there is an inevitable loss of resolution. Even if we use one of the newer-generation large-format CCD chips, there are few optical designs that have a sufficiently corrected field to take advantage of these large chips. Therefore there remains a very real role for the art of mosaics in producing large-format images while maintaining a high level of resolution.

Soon after I began astrophotography I became intrigued by the idea of splicing together multiple frames to achieve exciting large-format images. I started creating mosaics using a small refractor but, soon realized that, if I really wanted to explore the boundaries of amateur astro-imaging, I would need to use a longer-focal-length instrument. At this time I had purchased a 12.5-inch Ritchey-Chretien

Figure 10.2A. Grayscale mosaic.

Figure 10.2B. Registering the color data with Registrar.

Figure 10.2C. This high-resolution mosaic of M31 required stitching together 40 separate frames over three months for a total cumulative exposure of 50 hours. To create a color version, color data were taken from an older, lower-resolution image of M31 and registered to the grayscale mosaic using RegiStar. An LRGB composite was then made in Photoshop using traditional methods. The grayscale mosaic was made using a 12.5-inch RC at F9. The color data were taken with an ST10E and 4-inch refractor at F5.

Figure 10.2C. This high-resolution mosaic of M31 required stitching together 40 separate frames over three months for a total cumulative exposure of 50 hours. To create a color version, color data were taken from an older, lower-resolution image of M31 and registered to the grayscale mosaic using RegiStar. An LRGB composite was then made in Photoshop using traditional methods. The grayscale mosaic was made using a 12.5-inch RC at F9. The color data were taken with an ST10E and 4-inch refractor at F5.

Cassegrain with a prime focus of 3000mm. Using this instrument I was able to produce mosaics of much higher resolution and much greater image scale (see Figure 10.2 a-c), but this also required much more work and time stitching the many frames together as the field of view was contracted. These are truly laborintensive projects but the rewards are well worth the effort. The very large file sizes, generated by this process of combining multiple images, are sufficient for making very large-format prints. For me this is the most rewarding aspect of the process.

Probably the most critical element to success with large mosaics and other complex imaging projects is proper planning. I attribute any success I have had to proper planning. I often use programs like "The Sky" to help me plan the number and size of frames needed. I also make charts of the object using a template of the field of view of my imaging system. Preparation in this way is an essential exercise before going out with the telescope.

The art of creating mosaics can be found in detail on my Web site at http://www.robgendlerastropics.com/Article3.html and in Sky Telescope, June 1999, pp 138-141.

Composites. Composites are images made by layering separate frames taken of the same field using different techniques. Each frame is imaged using tech-

Figure 10.3A. High-resolution image of M81.

niques needed to optimize resolution and contrast for each part of the final image. The different frames are registered using RegiStar and then blended or layered together in Photoshop. The classic composite would be a short exposure of the trapezium layered over a deeper image of the Orion nebula. At about the same time I began creating mosaics, I also became interested in nontraditional composites. I asked myself why not substitute long-focal-length high-resolution images of selected areas, having abundant small-scale detail, onto a low-resolution background, which lacks small-scale detail and therefore does not require labor-intensive high-resolution imaging? In this way, the detail will be recorded where it is needed while other parts of the field that lacked small-scale detail would be spared the labor of high-resolution imaging. This makes for efficient but creative imaging (see Figure 10.3 a-c).

Hydrogen Alpha Color Composites. a natural evolution of hydrogen alpha (H-alpha) imaging of nebulae is the creation of color images using narrowband H-alpha data. The primary obstacle to incorporating H-alpha data into a color image is the unnatural "fit" of the narrowband data into the luminance of a traditional LRGB composite. Initial attempts to use H-alpha data as luminance were disappointing and often resulted in muted colors, monochrome-appearing nebulosity and bizarre-looking stars.

Figure 10.3B. High-resolution image of M82.

Figure 10.3C. High-resolution images of M81 and M82 (taken at a focal length of 3 meters) were layered on a background taken at 1 meter FL to achieve the final composite. The high-resolution components (Figures 10.3A and B) were taken using a 12.5-inch RC at F9 and the background was taken with an STL11000 and AP155 at F7. (Note the small satellite galaxy near M81 is Holmberg IX.)

Figure 10.3C. High-resolution images of M81 and M82 (taken at a focal length of 3 meters) were layered on a background taken at 1 meter FL to achieve the final composite. The high-resolution components (Figures 10.3A and B) were taken using a 12.5-inch RC at F9 and the background was taken with an STL11000 and AP155 at F7. (Note the small satellite galaxy near M81 is Holmberg IX.)

After quite a bit of experimenting I found that a natural-appearing image with aesthetic color balance could be obtained by incorporating the H-alpha data directly into the red channel of the RGB. I discovered that there were several ways to accomplish this successfully. One method was to make a 50/50 blend of the red channel and H-alpha data in Photoshop using the normal blending mode. The resulting image had much of the desirable contrast and signal of the H-alpha data along with the traditional-appearing stars of the red channel. I later modified this

Figure 10.4B.

H-alpha channel.

Figure 10.4B.

H-alpha channel.

Figure 10.4D. To make an H-alpha color composite of the Horsehead region I first created a blend of the red-filtered exposures and hydrogen alpha-filtered data using normal mode in Photoshop. After flattening this blend, I had a new red channel with which I made a traditional RGB. Cumulative exposure time was 12 hours. The background was taken using an AP155 refractor at F7. The high-resolution components were taken with a 12.5-inch RC at F9.

Figure 10.5B.

Low-resolution RGB image.

Figure 10.5B.

Low-resolution RGB image.

technique so that the H-alpha data can be used at 100% opacity in the blend. Using the red channel as the top layer one can then substitute the stars of the red channel into the H-alpha layer using the lighten mode at 100% opacity. Alternatively, if the previous method doesn't work well, the stars in the red channel can be selected using the color range (choose Highlights, then expand 1 to 3 pixels), inverse select and delete everything but the stars. Once the image is flattened it will have all the signal of the H-alpha component and the stars of the red channel. It may also help to first register the red channel with the H-alpha image using RegiStar since the data is often recorded on different nights and may need aligning. Once the new red channel is prepared it can be combined with the green and blue channels in a traditional way. Because of the intensity of the H-alpha/red channel the final RGB may need to have the "red" adjusted down. If needed, the H-alpha/red channel can be used as luminance in a traditional LRGB if this further enhances the image.

The final H-alpha color composite will have the remarkable depth of the hydrogen alpha image plus the beauty of a well-balanced color image (see Figure 10.4a-d).

Complex Hybrid Images

The natural evolution of these techniques is the construction of very complex images making use of all the methods described. Images of this nature are complex amalgamations of different resolution and color components and are often mosaics.

These types of imaging projects often require many hours of data collection and can take weeks or months to complete. The time and energy spent are almost always well worth it as the finished product is often a very unique and striking image and can be printed at tremendous sizes. (See Figure 10.5a-e and Figure 10.6.)

Figure 10.5E. For M33 a 15-frame grayscale mosaic was created first. This was done at long focal length (3000mm) using a 12.5-inch RC and ST10XME to produce a high-resolution image with FWHM of 2 arcseconds or less. A low resolution RGB image taken with a 4-inch refractor was used for the color component. A separate color image was made using three hours of hydrogen alpha-filtered data to highlight the numerous HII regions present in M33. A blend was then made of the conventional RGB and H-alpha color composite using the "Color range" tool in Photoshop. The HII regions were selected from the H-alpha color composite and layered into the conventional RGB. Next the color image was registered to the high-resolution grayscale mosaic in RegiStar. The final image is a high-resolution color mosaic highlighting the numerous HII regions of M33. Cumulative exposure time was 21 hours.

Figure 10.5E. For M33 a 15-frame grayscale mosaic was created first. This was done at long focal length (3000mm) using a 12.5-inch RC and ST10XME to produce a high-resolution image with FWHM of 2 arcseconds or less. A low resolution RGB image taken with a 4-inch refractor was used for the color component. A separate color image was made using three hours of hydrogen alpha-filtered data to highlight the numerous HII regions present in M33. A blend was then made of the conventional RGB and H-alpha color composite using the "Color range" tool in Photoshop. The HII regions were selected from the H-alpha color composite and layered into the conventional RGB. Next the color image was registered to the high-resolution grayscale mosaic in RegiStar. The final image is a high-resolution color mosaic highlighting the numerous HII regions of M33. Cumulative exposure time was 21 hours.

Figure 10.6. The Orion deep field is a mosaic of four separate frames. Each frame is a hydrogen alpha color composite with layered high-resolution components. The background image data were taken with a 4-inch refractor at F5 and STL11000 camera. The higherresolution components were taken with an ST10XME, AP155 and 12.5-inch RC. Cumulative exposure time was 20 hours.

Figure 10.7. Robert Gendler and his "observatory." Actually the telescope and mount are attached to a set of "wheelybars" (JMI), which is all wheeled out of the garage onto the driveway. The driveway is where all the imaging is carried out. This is a typical night with the southern sky filled with the "skyglow" of adjacent towns and neighborhood lights. The limiting visual magnitude on an excellent night is about 4.5 magnitude.

Was this article helpful?

0 0
Understanding Adobe Photoshop Features You Will Use

Understanding Adobe Photoshop Features You Will Use

Adobe Photoshop can be a complex tool only because you can do so much with it, however for in this video series, we're going to keep it as simple as possible. In fact, in this video you'll see an overview of the few tools and Adobe Photoshop features we will use. When you see this video, you'll see how you can do so much with so few features, but you'll learn how to use them in depth in the future videos.

Get My Free Video


Post a comment