Out of all the techniques employing biological particles, palynology is perhaps the one with the most varied range of applications. Palynology is a slow laborious process, involving a series of laboratory treatments to extract the pollen, followed by locating, identification and counting of pollen grains on microscope slides: in short (as put by Walker, 1990), ‘an enormous amount of work’.
In 1968, Professor Flenley discussed the ‘needs and prospects’ of automated palynology to significantly reduce the hours of labour needed in the palynology process.
The Pollen Classifynder is a digital microscope and software system that has been specifically developed as an aid to Palynology. The system enables slides to be scanned and the entire contents to be located, measured and the resulting digital images to be archived and then classified automatically.
The system requires minimum intervention from the user. The system also includes software that measures optical and texture parameters for every image captured. These parameters are used as the input to a neural network classifier that has been pre-trained to recognise the range of pollen expected on the slide or series of slides.
The system is based on optical microscopy and so is limited in its resolution by its optics and the wavelength of light. The current system uses darkfield illumination however this may change in the future as recent improvements in the intensity of the lighting system (based on light emitting diodes) has significantly improved available image contrast.