Soil optical properties affect spectral response of crop canopies and induce noise onto the relationships between reflectance data and crop characteristics such as leaf area index (LAI) or absorbed PAR (APAR). Different combinations of red and near infrared bands have been proposed but still suffered from a high sensitivity to soil brightness. As ideal vegetation index does not exist, we describe an improved vegetation index, the transformed soil adjusted vegetation index (TSAVI). It is based on similar principles as those stated by Huete (1988) and Major et al. (1988). We have used the SAIL model to simulate in different conditions the relationships between TSAVI and LAI or APAR. Experimental data recorded on wheat crops during the growing season are in good agreement with previous theoretical results.The W -TSAVI relationship can be set in the form oE TSAVI=TSAVI,(l-exp( -KTSAVI. LAI)) TSAVI, depends slightly on leaf angle distribution. The extinction coefficient, KTSAVI, depends on solar zenith angle and leaf angle distribution. The relationship between TSAVI measured at solar noon and MAR can be approximated to the linear simple form: MAR= c.TSAVIThis relationship is also in good agreement with experimental results with c = 1.205.

The detection of wheat heads in plant images is an important task for estimating pertinent wheat traits including head population density and head characteristics such as health, size, maturity stage, and the presence of awns. Several studies have developed methods for wheat head detection from high-resolution RGB imagery based on machine learning algorithms. However, these methods have generally been calibrated and validated on limited datasets. High variability in observational conditions, genotypic differences, development stages, and head orientation makes wheat head detection a challenge for computer vision. Further, possible blurring due to motion or wind and overlap between heads for dense populations make this task even more complex. Through a joint international collaborative effort, we have built a large, diverse, and well-labelled dataset of wheat images, called the Global Wheat Head Detection (GWHD) dataset. It contains 4700 high-resolution RGB images and 190000 labelled wheat heads collected from several countries around the world at different growth stages with a wide range of genotypes. Guidelines for image acquisition, associating minimum metadata to respect FAIR principles, and consistent head labelling methods are proposed when developing new head detection datasets. The GWHD dataset is publicly available at http://www.global-wheat.com/and aimed at developing and benchmarking methods for wheat head detection.

The soil line, a linear relationship between bare soil reflectance observed in two different wavebands. is widely used for interpretation of remotely sensed data. The basis on soil line was analyzed using a radiative transfer model in which reflectance was splitted into its single and multiple scattering components. The slope of the soil line corresponded to the ratio of the single scattering albedos corresponding to the two wavebands where the soil line was defined. The intercept originated from the difference in multiple scattering observed in each of the two wavelength bands used. The soil line concept was very robust over the whole optical domain as soon as soil types are separated, and when the effect of the view and source configurations as well as the surface roughness were considered. However, in the middle infrared spectral domain, the soil line concept failed when soil moisture was a factor of variation.