3.2. Discrimination algorithm
Fig. 11 shows a functional block diagram that details the data processing stages of the discrimination algorithm on a per scan basis, where, as illustrated, a single scan consists of:
1. Ambient light level measurement, with all lasers turned off.
2. 635 nm red laser reflectance measurement.
3. 685 nm red laser reflectance measurement.
4. 785 nm NIR laser reflectance measurement.
On a per scan basis, the individual measured reflectance levels were then adjusted by subtracting the measured background to produce three independent 512 element data vectors denoted by
! ! !
V635, V685 and V785. Using a defined peak threshold, Tpeak, peak
!
detection on the adjusted V785 values was performed. The peak
!
detection algorithm scanned the V785 data for values above Tpeak, performing two operations, namely, (1) single peak identification, representative of either a full leaf strike or ground strike, and (2) dual peak identification, representative of a partial leaf strike. For detected peaks corresponding to full leaf or ground strikes, the start pixel index, end pixel index and local maximum value were stored. Detected dual peaks corresponding to partial leaf strikes were flagged, as the peak information could not be used further by the discrimination algorithm. On completion, the output of the peak detection algorithm was a 1 30 vector denoted by
! !
V785PKs. Using the valid located peaks stored in V785 PKs the corre-sponding peak values in the V635, V685 and V785 vectors were extracted and the inferred target reflectance values calculated using measured BOP values stored in a lookup table (LUT).
3.2. Discrimination algorithmFig. 11 shows a functional block diagram that details the data processing stages of the discrimination algorithm on a per scan basis, where, as illustrated, a single scan consists of:1. Ambient light level measurement, with all lasers turned off.2. 635 nm red laser reflectance measurement.3. 685 nm red laser reflectance measurement.4. 785 nm NIR laser reflectance measurement.On a per scan basis, the individual measured reflectance levels were then adjusted by subtracting the measured background to produce three independent 512 element data vectors denoted by! ! !V635, V685 and V785. Using a defined peak threshold, Tpeak, peak!detection on the adjusted V785 values was performed. The peak!detection algorithm scanned the V785 data for values above Tpeak, performing two operations, namely, (1) single peak identification, representative of either a full leaf strike or ground strike, and (2) dual peak identification, representative of a partial leaf strike. For detected peaks corresponding to full leaf or ground strikes, the start pixel index, end pixel index and local maximum value were stored. Detected dual peaks corresponding to partial leaf strikes were flagged, as the peak information could not be used further by the discrimination algorithm. On completion, the output of the peak detection algorithm was a 1 30 vector denoted by! !V785PKs ใช้ยอดอยู่ถูกเก็บไว้ใน V785 PKs คอร์-sponding ค่าสูงสุดในเวกเตอร์ V635, V685 และ V785 ถูกสกัดและค่าแบบสะท้อนแสงสรุปเป้าหมายที่คำนวณโดยใช้วัด bop ที่มีระดับค่าเก็บไว้ในตารางการค้นหา (ลูฏ)
การแปล กรุณารอสักครู่..
