Image database: Apple ‘Golden’

Database of 100 calibrated images of apples’Golden’ (25 apples photographed by four sides). A reference to the acquisition system can be found in:

  • Vidal A, Talens P, Prats-Montalbán JM, Cubero S, Albert F, Blasco J (2013) In-line estimation of the standard colour index of citrus fruits using a computer vision system developed for a mobile platform. Food and Bioprocess Technology, 6(12), 3412-3419. DOI: 10.1007/s11947-012-1015-2.
  • Cubero S, Diago MP, Blasco J, Tardaguila J, Millán B, Aleixos N (2014) A new method for pedicel/peduncle detection and size assessment of grapevine berries and other fruits by image analysis. Biosystems Engineering. Special Issue on Image Processing in Agriculture, 117, 62-72. DOI: 10.1016/j.biosystemseng.2013.06.007.

The image acquisition system was composed of a digital camera (EOS 550D, Canon Inc., Japan) used to acquire high quality images with a size of 3456 x 2304 pixels and a resolution of 0.03 mm/pixel. The images were taken by placing each sample inside an inspection chamber in which contained the camera and the lighting system. The camera was placed at a distance of 20 cm from the samples. Illumination was achieved using four lamps that contained two fluorescent tubes each (Biolux L18W/965, 6500 K, Osram AG, Germany). The angle between the axis of the lens and the sources of illumination was of approximately 45º since the diffuse reflection responsible for the colour occurs at 45º from the incident light. However, the samples have a curved shape that can still produce bright spots affecting the colour measurements. To minimise the impact of these specular reflections cross polarisation was used by placing polarising filters in front of the lamps and in the camera lenses. The fluorescent tubes were powered using high frequency electronic ballast to avoid the flickering effect of the alternate current and produce a more stable light. Four images of each fruit were taken from two sides, stem and calyx. The apples were also weighed and measured (data contained in the excel file). The names of the files include the apple number and view. Figure 1 shows the vision system used to acquire the images.

The application Canon EOS utility was used to capture the images of the fruit. A standardised colour checker (ColorChecker SG Chart, X-Rite Inc, USA) was used to acquire colour references. Images of fruit were stored in JPG format due to internet limitations but images of colour checker were captured in both JPG and RAW (CR2) format to check the quality loss due to JPG compression. The results of the analysis of the colour patches in different colour spaces were completely the same with a correlation R2 higher than 99.9%

Download the here images (please, do not forget to acknowledge Cofilab when use it)


Contact for updates and support

Your Name (required)

Your Email (required)

Your Message

Uso de cookies

Este sitio web utiliza cookies para que usted tenga la mejor experiencia de usuario. Si continúa navegando está dando su consentimiento para la aceptación de las mencionadas cookies y la aceptación de nuestra política de cookies, pinche el enlace para mayor información.plugin cookies