Agricultura – Ştiinţă şi practică no. 3- 4(91-92)/2014 Agriculture - Science and Practice - 131 - SOME CONSIDERATIONS REGARDING THE USAGE OF MULTISPECTRAL REMOTE SENSING IMAGES IN AGRICULTURAL CROP ANALYSIS Pop N. 1) , T. Toderas 1) , Sanda Nas 1) 1) University of Agricultural Science and Veterinary Medicine, Cluj-Napoca, Manastur Street, No.3-5, 400372, Romania Abstract. The article includes some considerations regarding the usage of multispectral remote sensing imagery in agricultural crops analysis. Several aspects regarding the Landsat multispectral remote sensing image processing are taken into account, in order to perform the quantitative and qualitative anlysis on the cultivated vegetation. Keywords: Remote sensing, multispectral images, agricultural crops, quantitative and qualitative analysis INTRODUCTION Objectives: Two objectives were aimed: the efficiency and precision in identifying the agricultural crops classes from the researched area, as well as their qualitative analysis level. The precision in identifying the classification of the topographic details in a multispectral remote sensing image depends on the image’s spectral and spatial resolution. The spectral resolution refers to the spectral interval (wavelength interval) where an image was recorded and allows an object or phenomenon in the area to be identified through the electromagnectic radiation, reflected within an atmospheric window. [Jensen, J. R., 2005; Vlaicu Aurel, 1997]. In multispectral images, the spectral resolution is expressed through the number of bands (spectral intervals), where images of the same land surface were taken simultaneously. The spectral resolution depends on the spectral behavior of the topographic details in relation to the wavelength of the reflected electromagnetic radiation and the sensor’s sensitivity in relation to different spectral intervals. The spatial resolution refers to the linear dimension of the smallest topographic land detail that can be recorded in a remote sensing image. This corresponds as dimmension in land to the pixel edge of the image and depends on the performance of the remote sensing sensor [Vlaicu Aurel, 1997]. MATERIALS AND METHODS For the multispectral analysis of the vegetation from the researched area, Landsat multispectral images were used, especially those from the red (R) and near-infrared (NIR) bands. In the case study, three areas from the Landsat multispectral images were chosen, in June 2011, areas located in the unincorporated area of Câmpia Turzii locality. The image analysis process implied the image georeferencing process, then a quantitative and qualitative thematic analysis. The quantitative analysis was made through a thematic classification (both supervised and unsupervised), and for the qualitative analysis difference vegetation indices were used.