Multispectral Analyser Filters for Agricultural Monitoring
Within modern agriculture, increasingly large scales lead to a number of challenges. Vast fields make accurately monitoring and analysing crop growth across the area very difficult. Inefficient crop growth in certain areas can lead to reduced crop yields and higher field maintenance costs. For some time, the use of satellite imaging has been popular within large scale agriculture in order to build up a picture of the crop growth across large areas. This method, although more efficient than ground based visual monitoring still has a number of drawbacks; satellite images must be ordered in advance, and are only available once per day. Image resolution is relatively low and is highly dependent on the weather. Satellite imagery can be a useful for understanding the general uniformity of crop growth, but for accurately mapping areas of poor crop growth which may need attention, a more precise method is required.
In recent years the use of unmanned drones has become more prevalent in a number of industries. Agriculture is no exception. By using drones equipped with visible and NIR sensors, one can map the Normalised Difference Vegetation Index (NDVI) across the field. The NDVI is a measurement of the amount of live vegetation in an area. Chlorophyll is highly absorbent in the visible part of the spectrum, and highly reflective in the NIR. As the amount of chlorophyll present in a plant decreases, the amount of near infra-red light reflected decreases. The NDVI compares the reflected intensities of visible and NIR light. Areas with a high NDVI indicate areas with a high density of plant life containing chlorophyll, and thus a healthy crop. In contrast, a mid to low NDVI could indicate an unhealthy crop, or an area with no plant life at all. The data required to calculate NDVI is obtained through multi-spectral analysis, whereby images are captured at specific wavelength ranges across the electromagnetic spectrum, in this case; blue, green, red, and NIR.
Agricultural monitoring drones are commonly equipped with either CCD or CMOS sensing technology. The sensitivity of both CCD and CMOS sensors stretches from UV to near infra-red, so in order to use them for multispectral analysis, optical bandpass filters are required. LASER COMPONENTS, alongside partners Omega Optical, now offers a new range of filters specifically designed for remote monitoring of plant health and growing conditions. When combined with CMOS or CCD sensors, these filters enable high-precision data capture for processing and analysis using the NDVI.
A number of filters with standard wavelengths for this application are available on short lead times, with a thickness of 1.0 mm and can be cut to meet customer size specification. Designed specifically with multispectral analysis in mind, the series uses a single-surface coating in order to reduce secondary ghosting, filter weight and focus shift. A high effective index design allows the widest field of view possible with minimal spectral shift, and a minimised surface micro structure enhances resolution and pixel to pixel consistency. We are able to offer custom thicknesses for mounting filters directly on to CMOS or CCD sensors to maximise signal throughput by further eliminating reflections and ghost images. These filters are supplied with an optically matched adhesive to permanently bond the filter to your sensor.