BracketMerge is a digital image enhancement application used to merge exposure and/or focus bracketed images. BracketMerge enhances brightness, color saturation and depth of field. BracketMerge operates as an Image Based Enhancement tool. The information in multiple exposure and/or focus bracketed images is combined or blended to produce a result image that has improved brightness, color saturation and depth of field. The term Image Based Enhancement is used to imply that image enhancements are controlled by and even limited by the pixel information contained in the various input images. This Image Based Enhancement process produces result images with realistic brightness, color and focus characteristics.
Initial images are taken of the same scene with some form of exposure or focus bracketing. BracketMerge supports tiff and jpg formats. The best results are achieved for images captured and stored with 16 bits of data per color channel. BracketMerge supports images with either 8 or 16 bits per channel but in either case, computations are carried out with 16 bits of precision. BracketMerge merges or blends images two at a time. Each merge operation is referred to as a frame. Each frame will have two input, images referred to as the Base and Merge images, and will produce a single result image. A complete merge process that produces a single result image can consist of a single frame that merges two images or several frames merging several focus and exposure bracketed images. In the examples section of this article there are examples consisting of a single frame and as many as 14 frames. In the case of merge operations consisting of several frames, the result images from earlier frames become the input images in later frames.
As a first step in merging a pair of images BracketMerge will analyze the two images and align the pixels of one of the images in order that scene features
will be precisely matched up. This step is very important and must be done accurately. Once an alignment has been performed for a frame the Merge image can be saved in its aligned form. This is not necessary as part of the process of merging images. It is a capability that is provided to allow the use of precisely aligned images as input to other image editing applications. BracketMerge is going to merge the two images one pixel at a time. Pairs of pixels from the input images are combined in some way to produce one pixel in the output image. This may be as simple a process as determining which pixel in each pair of pixels is the best and using that pixel in the result. Or the process may consist of combining the properties of each of the pixels in each pair in proportions according to how much better one of the pixels may be in relation to the other. However you choose to combine your images, each pixel needs to be assigned a value. It will be this value that is used to determine which is the better of the two pixels, and by how much. BracketMerge lets you choose between the pixel properties of Contrast and Saturation in the determination of a pixel's value. Once pixels have been assigned values you can elect to average a pixel's value with the values of the pixels in its immediate neighborhood. This averaging, or smoothing operation, is particularly useful, actually necessary, when using contrast to merge images taken using focus bracketing. Saturation is the pixel property used when merging exposure bracketed images. Contrast is the pixel property used when merging focus bracketed images. Contrast is a measure of the amount of change between a pixel and its neighboring pixels. An image or image area with better focus will consist of pixels with greater contrast. In the first step BracketMerge aligned the images. In the second step each pixel was assigned a value. Step three will be to combine pairs of pixels to produce a result image. Controlling the blending operation has three parts:
* Pre Process: Determine the brightness of the result.
* Pixel Blending: Blend the color properties of pairs of pixels.
* Post Process: Adjust the amount of color saturation enhancement.
The Pre Process is optional, usually enabled when blending two images taken using exposure bracketing and usually disabled when blending two images taken using focus bracketing. In exposure bracketing one image will be brighter than the other. In this case you need to adjust the brightness of the pixels before they are blended together. BracketMerge lets you specify this in terms of the initial brightness of the two pixels in a pair. A value is chosen between -0.5 and 1.5. A value of 0.0 indicates brightness equal to the brightness of the darkest of the two pixels. A value of 1.0 indicates brightness equal to the brightness of the brightest of the two pixels. A value of 0.5, as you may be able to guess, indicates brightness half way between the brightness of the two pixels being merged. Values less than 0.0 and greater than 1.0 provide a way to produce a merged image that is either darker than the darkest of the two images or brighter than the brightest of the two images.
This optional pre process used all by its self provides a unique way to adjust the brightness in a photograph. In exposure bracketed images some pairs of pixels will vary quite a bit in brightness from one image to the next. Other pairs of pixels may be darker in both images or lighter in both images. Using this process to adjust pixel brightness will have little effect on pixels in pairs that have nearly the same brightness to begin with. Areas that are dark in both images will remain dark. Areas that are bright in both images will remain bright in the result image. The pixel blending step in the process can
be set up so that it will always take 0% of pixels from one image and 100% of pixels from the other image. If the brightness adjustment pre process was set up for instance to 1.25, the effect would be to brighten the brightest of the two images. But again, areas that were dark in both input images would remain dark in the result image. The pixels from the darker image are not merged with pixels from the brighter image, rather the difference in brightness between two images is used to control a brightness adjustment of the brighter image alone. Pixel Blending is controlled by specifying a function that determines the proportions of the input pixels to be used in producing a result pixel. For those with an aversion to all things mathematical there is no need to worry. The specification of this function is performed with a nice friendly graphical interface. A small set of slider widgets is used to affect a graph that represents the blending proportions.
The red/blue graph you see here represents the proportions that will be used when blending a pair of pixels. You see the terms "Base Image" and "Merge Image". This is the way that BracketMerge identifies the two images being blended. Here is how to interpret the meaning of the graph. Values were associated with every pixel of the two input images. The values associated with each pixel in a pixel pair determine a point along the horizontal axis. If the values for a pair of pixels are the same the proportions used to blend the pixels will be taken from the center of the graph. In this case it is a little hard to know what that is. Since the "Blending Function X Offset" and "Blending Function Y Offset" are set to 0 I can tell you that the graph is calling for blending proportions of 50% and 50%. In the case where the value associated with the "Base Image" pixel is larger than the value associated with the "Merge Image" pixel a greater and greater proportion of the "Base Image" pixel is used.
The license of this software is Free Trial Software, the price is $39.00, you can free download and get a free trial.