It depends on what you want to do with the file after you've saved it. If you want to maintain the "pixel values" from your float array, you can write out a binary file that contains the single- or double-precision float data. There is a write binary file function that will work for this.
If you want to be able to open the image in a graphics program then you will need to convert the image to either 8-bit or 16-bit integer data. Then decide whether you can live with a lossy compression (such as JPEG). If you can not live with any alteration of the image data, and you want to keep 16-bits per pixel, TIFF is a format that can support that. There is a IMAQ Write File 2 vi for that. (Select the TIFF format from the polymorphic VI).
Here in our lab we use a binary file format (of our own design) with an ASCII header and multiple images after the header. The header describes the image width, height, type (e.g. uint8, uint16, float32), big or little endian, grayscale or color, and so on. Anyone can open the file (with notepad for example) to figure out what is in the file. We have a reader VI that parses the header and knows how to read the pixel data.
We also have a plugin for ImageJ (freeware) that can parse the file and read in the image sequence.
