Original papersA practical approach to comparative design of non-contact sensing techniques for seed flow rate detection
Introduction
Appropriate and uniform seeds planting result in better germination of plants and yield enhancement by minimizing competition between plants for available light, water, and nutrients (Karayel and Ozmerzi, 2002). Proper placement of seed in the soil in sowing operation is highly desired for optimum growth and high productivity of any crop (Gürsoy, 2014). Achieving the desired plant population largely depends on the performance of the seeds planter (Murray et al., 2006). Approaching to proper and precise sowing rate per unit area requires development of an accurate sensing device for evaluating the seeding performance of the planters. In the literature, there are various studies on developing of appropriate sensing device to determine real sowing rate.
Machine vision sensing technique, including an image acquisition system and a software to realize the desired measures using some image processing algorithms, have been frequently applied by some researches. Karayel et al. (2006) and Navid et al. (2011) used a camera system for evaluating individual seed spacing. The performance of the camera system was compared with a sticky belt test stand as a reference in terms of seed spacing evaluation. Yongfang et al. (2011) and Akdemir et al. (2014) developed a detecting method based on machine vision to count the number of fallen seeds from planting machine in a determined area on a conveyor band by using image analyses techniques. An acoustic seed sensing system was also developed based on a rising voltage value that a microphone senses in each impact of seeds to a steel plate (Karimi et al., 2012, Karimi et al., 2015). In this system, seeds hit to the plate one by one and related seed spacing data is automatically calculated.
Recently, fiber sensor has been used to detect the seed flow. In this technique when the beam of light is blocked by falling seeds, their shade fall on the surface of the receiver causing a difference in intensity. Then amplifier translates the intensity difference into a signal of voltage. Al-Mallahi and Kataoka, 2013, Al-Mallahi and Kataoka, 2016 presented a new methodology to estimate the mass of grain seeds, which flow in the shape of clumps. The methodology used an off-the-shelf digital fiber sensor to detect the behavior of the clumps, including the length and the density of the clumps. Ding et al. (2016) installed a time interval measuring fiber-optical sensor to sense each passing seed in a seeding outlet of one precision drill. In the proposed system, each passing seed was converted to time-interval sequence grabbed by a time-capture interrupt of a microcontroller. Seeding time-interval sequences were processed to estimate the seeding performance indexes such as normal index, miss index, and multiple index.
Numerous studies have been conducted for developing an optoelectronic system with infrared sensing method. The major part of the optoelectronic systems was a photogate consist of several pairs of infrared light emitting diodes (LEDs) along with photo-detectors as receivers. The optoelectronic sensing technique is relying on reflection of the infrared radiation ray from the individual seed as they pass through the photogate (Kocher et al., 1998, Lan et al., 1999, Xia et al., 2010).
Deividson et al. (2014) suggested the use of the infrared sensor DF robot RB-DFR-49 connected to the microcontroller to determine the distance between individual fallen seeds in laboratory tests. Raheman and Kumar (2015) designed an infrared sensor based embedded system for sensing seed flow through the delivery tube of a seed drill. In this system, released seeds from the seed metering device passed through the band of radiation coming out from the two infrared LEDs. On the other side, the reflected radiation was realized by the infrared receiver. The developed embedded system could detect choking in a seed delivery tube giving a digital output of 0 and 1 for no flow of seeds and flow of seeds, respectively. Lu et al. (2017) employed a photoelectric sensor to monitor the status of the seed tube by converting quantities of seed tube status into voltage signals. The photoelectric sensor was comprised of 4 infrared LEDs and 4 infrared electric triodes that were uniformly arranged around the seed tube to monitor the whole zone of the cross section of the seed tube. This sensor was designed to directly determine three statuses of the seed tube, including normal seeds (normally passed) missing planting (no seeds passed), and blocked planting (seeds were blocked in the seed tube).
Cuhac et al. (2012) employed light dependent resistors as receiver components for seed flow measurement. They presented a real-time wireless seed flow monitoring system for seed drill implements. The seed flow estimation was determined through the use of the seed counting information acquired by light emitting diodes and light dependent resistors, which were installed on every pipe.
Appropriate seed sensing device must be able to sense the passage of the seeds at different flow rates. In a further step, the mass of delivered seeds in a specific time period should be estimated by processing acquired data from sensors. Utilizing a sensing device includes transmitter and receiver units, installed on the opposite side of each other in seeds delivering tube, seems to be a proper technique to sense passing seeds. The different seed flows during the planting operation make different response signals. According to the mentioned bases, it is conceivable that developed sensing system can estimate seed flow rate by processing signal information corresponding to seeds being planted. This study was carried out as a part of a research project with an ultimate objective of developing a seed drill performance monitoring systems. The study reported here seeks to make a practical approach to comparative design of non-contact sensing techniques for seed flow rate detection. Such that, findings would be the basis for the subsequent researches about developing a real-time seed flow monitoring system.
Section snippets
Seeds
In this study, wheat seeds with an average thousand seed mass of 35.4 g were used as experiment material. The seeds had elongated shape with an approximate average length (L) of 7.1 mm, width (W) of 3.6 mm and thickness (T) of about 2.8 mm. Approximate sphericity (Sp) was determined about 0.58 using the axial dimensions of seeds in (Mohsenin, 1986) formula:
Sensing method selection considerations
There were different methods for seed sensing like ultrasonic, microwave and proximity sensors. As a result of the comparatively
Data analysis
Data normalization was performed during the data preprocessing step to transform all acquired data to a common scale. The min-max normalizer linearly rescaled every feature to the [0, 1] interval using the following formula:where the data point i normalized between 0 and 1, the original value of each data point i, is a minima among all the data points, and . a maxima among all the data points. A practical example of normalized data of sensing units
Discussion
In the designed test apparatus to eliminate the possible trial errors and acquire individual pulse signals from an exact same seeds flow, three sensing units were aligned on one seed tube (Fig. 5). The seeds fall due to gravity, and their initial velocity can be ignored. Since the vertical distance between each sensor was low (about 5 cm), there wasn’t effective difference between passing the speed of seeds through the sensors. Such conclusion was also inferred from an early test which had been
Conclusion
IR sensing system is acceptably used to estimate seed mass flow rate from the metering system according to the strong linear relationship between the actual seed mass changes and the system acquired voltages (r = 0.87). Due to the obtained results from designed IR sensing unit, it seems that it is wisely to select IR sensing unit as an appropriate non-contact sensing technique for estimating of sowing rate in Seed Drills. The objective of subsequent work would be developing an estimation model
References (23)
- et al.
Estimation of mass flow of seeds using fibre sensor and multiple linear regression modelling
Comput. Electron. Agric.
(2013) - et al.
Application of fibre sensor in grain drill to estimate seed flow under field operational conditions
Comput. Electron. Agric.
(2016) - et al.
Two fruit counting techniques for citrus mechanical harvesting machinery
Comput. Electron. Agric.
(2009) - et al.
Laboratory measurement of seed drill seed spacing and velocity of fall of seeds using high-speed camera system
Comput. Electron. Agric.
(2006) - et al.
Review of techniques for the mass flow rate measurement of pneumatically conveyed solids
Measurement
(2011) - et al.
Development of an image analysis system for sowing machine laboratory tests
Agric. Mech. Asia Afr. Lat. Am.
(2014) - Cuhac, C., Virrankoski, R., Häninen, P., Elmusrati, M., Hööpakka, H., Palomäki, H., 2012. Seed flow monitoring in...
- et al.
Usage of the DF robot RB-DFR-49 infrared sensor to detect maize seed passage on a conveyor belt
Comput. Electron. Agric.
(2014) - et al.
Design and experiment of performance testing system of multi-channel seed-metering device based on time intervals
Trans. Chin. Soc. Agric. Eng.
(2016) - Evans, J.D., 1996. Straightforward Statistics for the Behavioral Sciences....
Performance evaluation of the row cleaner on a no-till planter
Trans. ASABE
Cited by (45)
Line laser based sensor for real-time seed counting and seed miss detection for precision planter
2023, Optics and Laser TechnologyAn efficient seeding state monitoring system of a pneumatic dibbler based on an interdigital capacitive sensor
2023, Computers and Electronics in AgricultureA signal output quantity (SOQ) judgment algorithm for improving seeding quantity accuracy
2022, Computers and Electronics in AgricultureCitation Excerpt :The photoelectric monitoring method was the most widely used because of its low cost and high sensitivity. Karimi et al. (2017) conducted comparative experiments on three non-contact sensor monitoring methods of photoresistor (LDR), infrared (IR), and laser diode (LD). Their common principle was that when the seeds pass through the seed sensor and the band of light beams, their shades fall on the receiver elements, resulting in output voltage changes.
Design of smart seed sensor based on microwave detection method and signal calculation model
2022, Computers and Electronics in AgricultureCitation Excerpt :The indirect incidence TSOP based IR sensor comprised two IR LEDs to provide radiation and one TSOP based IR receiver to receive the reflected radiations. Karimi et al. (2017) conducted a comparative test on three photoelectric types of sensors: photoresistor (LDR), infrared (IR), and laser diode (LD). Test results showed that there was a strong linear relationship (r = 0.87) between the actual seed mass changes and the corresponding voltages of the IR sensing unit.
Design and testing of novel seed miss prevention system for single seed precision metering devices
2022, Computers and Electronics in AgriculturePrecision seeding parameter monitoring system based on laser sensor and wireless serial port communication
2021, Computers and Electronics in Agriculture