The ability to predict crop yield is beneficial and one being studied to provide added insight for farmers and researchers. In a study conducted by the University of Florida, an off-the-shelf Canon 10 MPixel camera with a 1/2.3″ CCD image sensor and USB interface captured images of citrus under different light conditions and of fruit hidden by leaves and branches in order to provide a more accurate estimate for crop yields.
A basic shape analysis enabled researchers to find tentative locations of citrus fruit hidden from initial view. Since most of the fruit typically has a circular shape, the algorithm was used to determine the parameters based on approximate size. The camera was trained to classify & identify the fruit using two types of features: local texture features and Tamura texture (coarseness, contrast, and directionality) features. In total, 10 features were chosen to define the local structure of the surface. After further enhancement, the algorithm was able to detect additional features to help researchers determine whether the image represented a fruit or a leaf.
Universe Optics can help scientists and researchers with specific lens needs for studies such as the one conducted by the University of Florida.
We carry over 1600 different lenses in stock at our New York facility with standard and custom lens assemblies for scanners, CCTV, CCD/CMOS, medical imaging, surveillance systems, machine vision and night vision systems. We also offer other value-added services such as special coatings, custom apertures, custom packaging, and anti-vibration assembly techniques.