The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-3/W6
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W6, 269–275, 2019
https://doi.org/10.5194/isprs-archives-XLII-3-W6-269-2019
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W6, 269–275, 2019
https://doi.org/10.5194/isprs-archives-XLII-3-W6-269-2019

  26 Jul 2019

26 Jul 2019

CROP INVENTORY OF ORCHARD CROPS IN INDIA USING REMOTELY SENSED DATA

K. Chaudhari1, N. Nishant2, G. Upadhyay1, R. More1, N. Singh1, S. P. Vyas1, and B. K. Bhattacharya1 K. Chaudhari et al.
  • 1Space Applications Centre, ISRO, Ahmedabad, Gujarat – 380 015, India
  • 2North Eastern Space Applications Centre, ISRO, Umium, Meghalaya – 793 103, India

Keywords: Orchard Crops, Mango, Banana, Citrus, ISO-DATA Clustering, Object oriented Classification

Abstract. The use of satellite remote sensing (RS) technologies for purpose of crop discrimination, mapping, area estimation, condition and yield assessment has been proved to be effective and efficient in terms of time and cost, having better consistency implemented with scientific approaches. However, application of satellite RS technology for horticultural crops in India has certain challenges due to scattered and small field sizes, comparatively short duration such as vegetable crops and mixed cropping. Hence the study was taken for developing research methodology for area assessment of three major fruit crops such as Banana, Mango and Citrus over 20 districts in four states viz. Gujarat, Madhya Pradesh, Uttar Pradesh and Bihar. Appropriate bio-window for analysing different crop types was selected and mapping of crops were done using pixel based hybrid classification i.e. un-supervised ISODATA clustering plus supervised MXL classification as well as object based classification of high resolution remote sensing data (Resourcesat LISS III and/or LISS IV, Cartosat – 1 PAN) followed by their accuracy assessment and their comparison with departmental reported statistics. Overall, the classification accuracy was more than 80% for all the crops. Deviation from statistics were in the range of 3 to 38%. Higher deviations from statistics were mostly due to use of lower resolution satellite data or mixing of crops having similar spectral signatures e.g. mango and sapota in Navsari and Valsad districts of Gujarat. It was very difficult to discriminate the young orchards of 2–3 years from other field crops due to mixed / inter cropping practices. The maps were checked and certified by respective State Horticulture Departments and were archived in VEADS, SAC and BHUVAN, NRSC geoportals of ISRO. RISAT – 1 (microwave) data were explored for the estimation of banana orchards in order to detect banana plantation at early stage and under cloudy sky conditions. There is huge potential of application in this sector using advanced observations from hyperspectral, thermal infrared sensors and advanced radars or LIDAR’s on-board upcoming satellites.