WO2019043356A1

WO2019043356A1 - Improvements relating to crop harvesting

Info

Publication number: WO2019043356A1
Application number: PCT/GB2018/052339
Authority: WO
Inventors: Keith Lawrence GEARY
Original assignee: LLEO Limited
Priority date: 2017-09-04
Filing date: 2018-08-17
Publication date: 2019-03-07
Also published as: GB2557469A; GB201721785D0; GB2557469B; GB201714167D0

Abstract

A method of agricultural harvesting is provided, the method comprising steps of obtaining one or more images of a crop region, assigning a plurality of crop zones to the one or more images of the crop region, measuring from the one or more images a parameter representative of crop growth for each crop zone, assigning one of a plurality of pre-determined grades to each crop zone, the grade for each crop zone being determined by the growth parameter for that crop zone, and harvesting the crop by separating and at least temporarily storing crops according to the grades assigned to the crop zones.

Description

Title - Improvements relating to Crop Harvesting

The present invention relates to methods and apparatus for harvesting crops, for example the harvesting of potatoes.

Potato is a major worldwide crop, which is required throughout the year by processors and consumers. Due to the seasonal growing of potatoes, long-term storage of potatoes is necessary, during which the harvested potato tubers must be maintained with a quality sufficient for processing and then consumption by consumers.

Following the harvesting of potato tubers, the potato tubers must be stored. The nature of the potato tubers is that they will commence a new vegetation cycle under warmer temperature and higher humidity, ie by sprouting. Unfortunately, the sprouting of the potato tubers often starts in storage, which results in

compositional changes, such as increased sugar levels, and which damages the potato tubers. The potato tubers lose quality and weight, which may ultimately result in the disaggregation of the potato tubers. At present, multiple strategies are used to extend dormancy, and minimise sprouting and waste, including the application of sprout suppressants, such as chlorpropham. Development of viable alternative strategies to maintain potato tubers and bulbs in a dormant state and to achieve long-term suppression of sprouting are key industry priorities. Indeed, long-term storage of potato tubers is essential for year-round supply. Maintaining sprout suppression and low reducing sugars during storage is essential for supply quality and minimising the formation of acry!amide; key priorities for the processing industry. Potato storage is still heavily reliant on the chemical suppressant, eg chlorpropham (CIPC), to manage sprouting, but many countries are considering or implementing restrictions on their use.

Furthermore, the selection of potatoes on their dry matter content or solids is very important to most processors. A higher dry matter content is of great importance because a potato tuber needs to accumulate a high concentration of dry matter to avoid extensive fat absorption during processing into products such as French fries and potato crisps. Dry matter concentration is highly correlated with tuber specific gravity, and the oil content of potato crisps decreases with increase in specific gravity. Specific gravity may also modify potato tuber quality attributes related to the flavour, odour, microorganism deterioration power, enzymatic activity, and tuber fry colour. The cost of frying oil is a major component in processing costs and it is therefore an aim of the processor to obtain potatoes with optimum dry matter levels, so as to maximise processed yield and to minimise oil costs.

There has now been devised a method of harvesting and apparatus for harvesting, which overcome or substantially mitigate the aforementioned and/or other disadvantages associated with the prior art.

According to a first aspect of the invention, there is provided a method of agricultural harvesting, which method comprises the steps of:

(a) obtaining one or more images of a crop region,

(b) assigning a plurality of crop zones to the one or more images of the crop region,

(c) measuring from the one or more images a parameter representative of crop growth for each crop zone,

(d) assigning one of a plurality of pre-determined grades to each crop zone, the grade for each crop zone being determined by the growth parameter for that crop zone, and

(e) harvesting the crop by separating and at least temporarily storing crops according to the grades assigned to the crop zones.

According to a further aspect of the invention, there is provided a crop harvesting apparatus comprising an imaging device for obtaining one or more images of a crop region, a data processor for assigning a plurality of crop zones to the one or more images of the crop region, measuring from the one or more images a parameter representative of crop growth for each crop zone, and assigning one of a plurality of pre-determined grades to each crop zone, the grade for each crop zone being determined by the growth parameter for that crop zone, and a crop harvesting machine for harvesting the crop by separating and at least temporarily storing crops according to the grades assigned to the crop zones.

The method and apparatus according to the invention are advantageous

principally because the measurement of a parameter representative of crop growth for each crop zone, and then assigning one of a plurality of pre-determined grades to each crop zone, the grade for each crop zone being determined by the growth parameter for that crop zone, enables a crop to be harvested by separating and at least temporarily storing crops according to the grades assigned to the crop zones. This may therefore enable crops with particular characteristics, such as quality, storage stability, nutritional content, dry matter content, etc, to be sorted according to those characteristics during harvesting, using pre-determined grading data.

At least temporarily storing crops according to the grades assigned to the crop zones may comprise storing crops on the crop harvesting apparatus according to the grades assigned to the crop zones and/or storing crops in long-term storage according to the grades assigned to the crop zones. In any one harvesting, at least one grade assigned to the crop zones may mean that the crops are stored in long- term storage and/or at least one grade assigned to the crop zones may mean that the crops proceed straight to processing.

The method of the present invention may be advantageous in respect of a wide variety of different crops. However, a crop with a particular problem that is overcome or substantially mitigated by the present invention is potato. In particular, development of viable alternative strategies to maintain potato tubers and bulbs in a dormant state and to achieve long-term suppression of sprouting are key industry priorities.

Crops generally emerge, grow to full size, mature and then die back (senescence), at different rates, even when the plants are from the same seed stock, planted at the same time, in the same field and tendered to in the same manner. Our research has shown that, in any harvest of potatoes, those potatoes taken from plants with tubers that matured later were more stable in storage than potatoes taken from plants, from the same area of the crop, with tubers that matured earlier. This correlation means that it is possible to identify which potatoes are likely to require a dose of chemical suppressant, eg CIPC, to manage sprouting and those that are not, meaning they could be placed in separate stores or different parts of a store. Another advantage is that tubers that matured earlier that are placed into storage maintain a lower level of sugars, which has been found to result in a better crisp when the potato slices are fried, as the hot oil caramelises the sugars that are present in the potato; too much sugar means that there will be a dark caramel present on the crisps.

Our research has shown that, in any harvest of potatoes, those potatoes taken from plants that emerge earlier have a higher dry matter content than potatoes taken from plants, from the same area of crop, that emerge later. Our research has also shown that, in any harvest of potatoes, those potatoes taken from plants that senesce later have a lower dry matter content than potatoes taken from plants, from the same area of crop, that senesce earlier. These correlations mean it is possible to identify which potatoes are likely to absorb more fat during processing, and which potatoes are likely to require longer frying time. These correlations also mean it is possible to harvest potatoes into batches that have a similar dry matter content, thus improving the efficiency of processing, for example by tailoring the frying time of the potatoes according to the time at which the batch's plants senesced.

In the step of obtaining one or more images of a crop, the image(s) may be collected by any suitable imaging device, eg an image sensor. The image data may be generated from at least some visible wavelength light (eg 400-600nm), eg red light (approximately 650-700nm), and/or at least some non-visible wavelength light (eg greater than 700nm), eg near-infrared light (eg greater than 700nm). The image data collected will depend on the parameter to be measured from the one or more images. The image data may be collected by a visible spectrum image sensor, such as a conventional digital camera, eg detecting wavelengths of less than 700nm, in combination with a near-infrared image sensor, eg detecting wavelengths of greater than 700nm, or alternatively by a multispectral image sensor that detects wavelengths of less than 700nm and wavelengths of greater than 700nm, eg at least 600-800nm.

The imaging device may be a CCD or a CMOS image sensor, and may have sufficient resolution to provide one or more images in which each plant is individually discernible, eg enabling the plants to be counted.

The imaging device may be a mu!ti-spectral camera, which may generate image data from a plurality of wavelength bands, for example any combination of a blue band (eg including at least some wavelengths in the range 455-495nm), a green band (eg including at least some wavelengths in the range 540-580nm), a red band (eg including at least some wavelengths in the range 660~680nm), a red- edge band (eg including at least some wavelengths in the range 710-730nm) and a near-infrared band (eg including at least some wavelengths in the range 800- 880nm). However, an imaging device comprising a greater number of bands, eg greater than 100 bands, such as a so-called hyper-spectral camera, may also be used.

The apparatus may also include a transmitter of electromagnetic radiation, and the imaging device may detect a reflected portion of the transmitted electromagnetic radiation. The one or more images may therefore consist at least partially of data generated from a reflected portion of the transmitted electromagnetic radiation that is detected by the imaging device. The transmitter of electromagnetic radiation may be a laser, or may form part of a radar system. Indeed, it will be appreciated that the image may be generated from non-visible electromagnetic radiation, such as radio waves. Examples of suitable systems include a LIDAR system (Light Detection And Ranging) and an SAR system (Synthetic-Aperture Radar).

The imaging device may be fixed relative to the crops, eg on a fixed boom or mast. Alternatively, the imaging device may be movable relative to the crops, eg on a vehicle.

The vehicle may be grounded, eg a car, a tractor, or a piece of machinery that is treating or harvesting the crops. Alternatively, the vehicle may be airborne, eg an aircraft or a remotely piloted aircraft system (ie a drone). The drone may be a fixed wing, single-rotor or multi-rotor drone. For some imaging devices, such as radar systems, eg an SAR system (Synthetic-Aperture Radar), location on a satellite may be suitable. The imaging device may be pre-installed on the vehicle. Alternatively, the imaging device may be retrofitted to the vehicle.

The image data may be transmitted to a CPU located on the vehicle, or where the vehicle operator is located externally from the vehicle, to an external CPU. The images may be recorded at predetermined time intervals as the imaging device is moved relative to the crops. The predetermined time intervals may be every millisecond, every second, or every two seconds.

Each of the one or more images may comprise a single field, a plurality of fields, or parts of one or more fields. The one or more images may each comprise a continuous image of a crop region, or alternatively a discontinuous image of a crop region, eg divided into discrete measurement zones, which may be partially or wholly separated from at least another measurement zone. The one or more images may comprise an array of pixels, each having data regarding intensity of light received by the image device at one or more wavelengths. A pixel of the image may contain data in respect of a single plant or a plurality of plants.

The one or more images of a crop may be obtained at a single time, or one or more images may be obtained at each of a plurality of times, for example an image may be obtained at pre-determined intervals over a time period. The time period may be sufficient to encompass at least the emergence of the plants from the ground. The time period may be sufficient to encompass maximum growth of the crops, and may also be sufficient to encompass reduction of the plants, eg die- back (senescence). The time period may be at least 1 week, at least 2 weeks, at least 1 month, at least 2 months or at least 3 months. The interval between successive images being obtained may be at least 1 day, at least 2 days, at least 3 days or at least 1 week. The time intervals may be regular, or substantially regular (less than 20% difference between maximum and minimum intervals).

A plurality of crop zones are assigned to the one or more images of the crops. The crop zones may comprise a plurality of zones within an area for harvest, which may correspond to an area that would be harvested in a single operation, eg by a harvesting machine. The area for harvest may therefore comprise one or more rows of plants, eg at least 2, at least 3, or at least 4 rows of plants. The crop zones may comprise an array of zones within each area for harvest. The crop zones may comprise a regular array of zones within an area of harvest, eg a rectangular array. The crop zones may be arranged in an area for harvest in a single row, with each crop zone extending transversely across the full extent of the area for harvest. In these embodiments, each crop zone may encompass a plurality of plants. Alternatively, each crop zone may encompass a single plant. The crop zones may be uniform in area, or may be uniform with respect to the number of plants present, or expected to grow, in each crop zone. A parameter representative of crop growth (the crop growth parameter) for each crop zone is measured from the one or more images. Where a crop zone comprises a plurality of plants, the crop growth parameter may be an average for the plurality of plants within the crop zone. The crop growth parameter may be selected to represent a growth characteristic, where a particular correlation between a growth characteristic and at least one characteristic of the harvested crop has been identified. The crop growth parameter may be representative of the maturity of the one or more plants in the crop zone. The crop growth parameter may be indicative of the number, size and/or health of the one or more plants in the crop zone. The crop growth parameter may correspond to a count or estimate of the number of plants within each crop zone. The crop growth parameter may correspond to a measurement of the area covered by the one or more plants in each crop zone, eg the canopy area, which may be an absolute measurement or a relative measurement with respect to the total area of the crop zone.

The crop growth parameter may be the Normalised Difference Vegetation Index (NDVI) of the crop zone. In particular, the NDVI may be calculated as follows: NDVI = (NIR - VIS) / (NIR + VIS) where VIS and NIR are the spectral reflectance measurements acquired in the visible (eg red) and near-infrared regions, respectively. An imaging device using a conventional RGB image sensor along with a near-infrared sensor, or using a multispectral sensor capturing light in at least the 600 to 800nm range, are effective in providing the data to enable calculation of the NDVI.

The image may be obtained at a particular time, such that the image data obtained for each crop zone may represent the same time of collection. However, particularly where an aircraft or a remotely piloted aircraft system (ie a drone) is used, the image data may be collected at different times in different parts of the image. However, the image data collected for the crop zones may be collected within a time period that is significantly less than the time period between successive images being collected, eg a time period that is less than 10%, or less than 5% or less than 1 % of the time period between successive images being collected. The image data collected for the crop zones may be collected within a time period of less than 3 hours, less than 1 hour, or less than 30 minutes, and the time period between successive images being collected may be at least 1 day, at least 2 days, at least 3 days or at least 1 week.

The crop growth parameter may vary over time, and the crop growth parameter may vary over the time period for collection by increasing from an initial value, which may be zero, and increasing to a maximum value. The crop growth parameter may then decrease either partially, or fully to the initial value. A crop growth profile, representing the change in the crop growth parameter over a time period, may therefore be measured. Where appropriate, the crop growth parameter that is measured may be a time, eg date, for example at which a plant (or an average of a plurality of plants) reaches a particular stage of growth or retreat, eg a percentage of growth or senescence. For example, the crop growth parameter may be the time at which a plant (or an average of a plurality of plants) reaches a pre-determined senescence, where the pre-determined senescence may be a percentage senescence in the range 10% to 90%, 20% to 80% or 30% to 70%.

The measured crop growth parameters may be stored in a suitable data file, eg a spreadsheet or a CSV file, for analysis by a separate system, eg software, for assigning grades to each crop zone, or this analysis may be performed by the same system, eg software, that processes the crop growth parameter

measurements.

One of a plurality of pre-determined grades is assigned to each crop zone, the grade for each crop zone being determined by the growth parameter for that crop zone. The measured crop growth parameters may be directly correlated to the plurality of pre-determined grades. In particular, each pre-determined grade may be assigned to a crop zone in which the measured crop growth parameter falls within a particular range. The ranges of measured crop growth parameters that correspond to the pre-determined grades may together form a continuous range of crop growth parameters. In addition, the pre-determined grades may comprise a maximum grade that corresponds to a measured crop growth parameter above (or below) a threshold value, and the pre-determined grades may comprise a minimum grade that corresponds to a measured crop growth parameter below (or above) a threshold value.

Alternatively, the measured crop growth parameter for each crop zone may take the form of a crop growth profile, ie a series of crop growth parameters measured at different times. In these embodiments, the plurality of pre-determined grades may each have a representative crop growth profile, against which the measured crop growth profiles are fit. The grade for each crop zone may then be determined by the representative crop growth profile against which the measured crop growth profile best fits.

The pre-determined grade that is assigned to each crop zone may correspond to a particular plant, where the crop zone encompasses a single plant, or may correspond to an average for a plurality of plants within the crop zone, where the crop zone encompasses a plurality of plants.

The pre-determined grades may be indicative of a characteristic of the harvested crop, and may be indicative of the quality of the harvested crop. The quality of the harvested crop may include storage stability, physical or chemical composition, appearance, or any other characteristic that is desired of the harvested crop. For potato tubers, in particular, the grade may be indicative of the storage stability and/or the sugar content and/or the dry matter content.

The crop is harvested by separating and at least temporarily storing crops according to the grades assigned to the crop zones. The crop zone may comprise a single plant or a plurality of plants. The crops from each crop zone may be transferred from the ground or the plants into a container for a particular grade, or a particular group or range of grades. The crop may be harvested by a crop harvesting machine, which may include a plurality of containers, each container being for receiving a particular grade, or a particular group or range of grades, of the crop. The crop harvesting machine may comprise a vehicle, and a mechanism for removing the crop from the ground or from the plants. The containers of the crop harvesting machine may be carried by a separate vehicle, or may be integrated into the vehicle carrying the mechanism for removing the crop from the ground or from the plants. The crop harvesting machine may include a system for identifying the location of the crop being harvested. This system may utilise satellite data for identifying location, eg GPS data, or this system may utilise information derived from a plurality of location devices within or around the crops being harvested. The crop harvesting machine may be operated to harvest crops from each crop zone in turn, one after another, with a sorting mechanism causing the crop from each crop zone to be transferred to an appropriate container, according to grade. This operation may be computer assisted or computer controlled, and may utilise one or more data files containing data regarding the location of the crop zones and the pre- determined grade that has been assigned to each crop zone. Where this operation is computer assisted, the crop harvesting machine may display crop zone location and grade data to the user, or may display only grade data that corresponds to a detected location to the user, to enable the user to select the relevant container for the sorting mechanism. Where this operation is computer controlled, the crop harvesting machine may control the sorting mechanism without the need for user input, with the relevant container being determined by use of crop zone location and grade data.

The crop harvesting machine may comprise means for Windrow harvesting. The crop harvesting machine may be operated to harvest crops from adjacent rows, one after another, with a sorting mechanism causing the crop from a first row to be deposited on to a second row of crops, wherein the crops in the first and second rows have been assigned the same grade. The crop harvesting machine may then harvest the crops of the first and second rows together. Alternatively, an additional machine may harvest the crops of the first and second rows once the crops of the first row have been deposited onto the second row of crops. The additional machine may also comprise a plurality of containers, and a sorting mechanism causing the crop from each crop zone to be transferred to an appropriate container, according to grade. This operation may be controlled similarly to the operation of the sorting mechanism of the crop harvesting machine described above. The software utilised by the crop harvesting machine may be a Geographic

Information System (GIS Software).

Once the crops have been harvested, the crops may be transferred to long-term storage. The separated, graded crops may be stored separately, and may be treated differently, eg with respect to physical or chemical treatment. For example, the separated, graded crops may have different storage temperatures, humidity, atmosphere, or applied chemicals. For example, for potato tubers, the most stable grade or grades of harvested crops may be treated with less, or no, suppressant chemical, such as CIPC. The separated, graded crops may also proceed to different processing and/or different products, dependent on the characteristic on which the crops were graded. Alternatively, some graded crops may proceed to processing immediately, and not be stored at all. A practicable embodiment of the present invention will now be described with reference to the accompanying figures, of which:

Figure 1 shows a series of multispectral images of a group of crop fields, which indicate the Normalised Difference Vegetation Index value (NDVI) for the crops, with the images being generated at weekly intervals from left to right;

Figure 2 shows an NDVI map of a group of crop fields, and a corresponding map in which the group of crop fields is divided into harvesting regions for which an average NDVI value is calculated;

Figure 3 shows an image of a group of crop fields, which has been divided into a plurality of numbered blocks;

Figure 4 shows a map of the group of fields of Figure 3, which has been divided into an alternative arrangement of numbered blocks;

Figures 5a-5d show a series of multispectral images of a portion of a group of crop fields, with the images being generated on 6th August (Figure 5a), 12^th August (Figure 5b), 16^ih August (Figure 5c) and 23^rd August (Figure 5d), and with Figure 5d identifying three crop zones A, B and C;

Figure 6 shows a table containing canopy, NDVI and population estimate data for crop zones A, B and C, across a series of dates from 31 May 2016 to 29 August 2016;

Figure 7 is a graph showing canopy coverage values for crop zones A, B and C as a function of time;

Figure 8 is a graph showing NDVI values for crop zones A, B and C as a function of time;

Figure 9 is a graph showing population estimate values for crop zones A, B and C as a function of time;

Figure 10 is a chart showing sprout break after storage of potato tubers grown in crop zones A, B and C; Figure 1 1 is a chart showing the solids within potato tubers grown in crop zones A, B and C, at intake and after 21 weeks;

Figure 12 is a graph showing canopy coverage values for crop zones A, B and C of a second sample of crops, as a function of time, taken across a series of dates from 10 June 2017 to 16 September 2017;

Figure 13 is a graph showing NDVI values for crop zones A, B and C of the second sample of crops, as a function of time, taken across a series of dates from 10 June 2017 to 16 September 2017;

Figure 14 shows an example harvesting map indicating the distribution of crop zones A, B and C across the fields of the second sample of crops; Figure 15 shows a table containing dry matter content for three samples within the crop zones A, B and C of the second sample of crops;

Figure 16 shows a multispectral image of a portion of a group of crop fields, with harvesting areas and graded crop zones indicated;

Figure 17 is a graph of filter transmissivity for a multi-spectral camera and typical plant reflectance, as a function of the wavelength of light; and Figure 18 shows a series of multispectral images of a group of crop fields, which have been processed on the basis of different crop reflectance indices.

In a practicable embodiment of the present invention, one or more images of a crop are obtained. These images are obtained using a drone with a 5 band multispectral camera, which includes visible Red, Green and Blue channels for generating an image from visible light, and a near-infrared channel for obtaining data from which the NDVI of the crops in each pixel of the image may be calculated. Figure 1 shows a series of multispectral images of a group of potato crop fields, which indicate the Normalised Difference Vegetation Index value (NDVI) for the crops, with the images being generated at weekly intervals from left to right. These images show how the NDVI changes over time for crops in different areas of the fields shown. Similarly, Figure 2 shows an NDVI map of a group of potato crop fields, and a corresponding map in which the group of crop fields is divided into harvesting regions for which an average NDVI value has been calculated.

Figure 3 shows an image of a group of crop fields, which has been divided into a plurality of numbered blocks. Figure 4 shows a map of the group of fields of Figure 3, which has been divided into an alternative arrangement of numbered blocks. These blocks represent areas of the fields that may then be divided into harvesting regions, which may then be divided into crop zones for which growth parameters may be measured. Figures 5a-5d show a series of multispectral images of a portion of a group of potato crop fields, with the images being generated on 6th August (Figure 5a), 12^th August (Figure 5b), 16^ih August (Figure 5c) and 23^rd August (Figure 5d), and with Figure 5d identifying three crop zones A, B and C. These plots represent crop zones that, in practice, would be distributed as a regular array across the full extent of the crop fields. However, these three plots have been chosen to illustrate the method according to the invention.

From the multispectral image data, values for the following parameters were calculated:

(a) Canopy = the proportion of the crop zone A, B or C that is covered by the crop

(b) NDVI - (NIR - VIS) / (NIR + VIS), where VIS and N!R are the spectral

reflectance measurements acquired in the visible (red) and near-infrared regions, respectively, with the NDVI value being the average for the crop zone A, B or C

(c) Population estimate = count of the number of plants in the crop zone A, B or C

Figure 8 shows a table containing the canopy, NDVI and population estimate data for crop zones A, B and C, across a series of dates from 31 May 2016 to 29 August 2016, as indicated in the table. From this data, the crop growth profiles shown in Figures 7 to 9 were generated. In particular, Figure 7 is a graph showing canopy coverage values for crop zones A, B and C as a function of time, Figure 8 is a graph showing NDVI values for crop zones A, B and C as a function of time, and Figure 9 is a graph showing population estimate values for crop zones A, B and C as a function of time. As shown in these graphs, the crops in crop zone A grow earlier than the crops in crop zone B, which in turn grow earlier than the crops in crop zone C. It has been found that the later the growth of the potato crop, the greater the storage stability of the harvested crop. Figure 10 is a chart showing sprout break after storage for potato tubers grown in crop zones A, B and C. As shown in this chart, the potato tubers grown in crop zone C show the longest storage time without sprouting, and the potato tubers grown in crop zone A show the shortest storage time without sprout break.

Figure 1 1 is a chart showing the solids within potato tubers at intake and at 21 weeks after intake, in crop zones A, B and C. As shown in this chart, the potato tubers grown in crop zone C show the best control of solids. This may lead to the potato tubers having a reduced bruising potential.

The crop growth profiles shown in Figures 12 and 13 were generated from canopy and NDVI data generated from an average of three samples taken within three different crop zones (A1 -A3, B1 -B3 and C1 -C3) across a series of dates from 10 June 2017 to 16 September 2017. In particular, Figure 12 is a graph showing average canopy coverage values for crop zones A ("Shelford-ear!y), B ("Shelford- mid") and C ("Shelford-late") as a function of time, and Figure 13 is a graph showing NDVI values for crop zones A ("Shelford-early), B ("Shelford-mid") and C ("Shelford-late") as a function of time. The crop zones are labelled on the graph according to the type of crop, eg "She!ford" potato tubers, and the stage at which senescence occurs, eg "late", "mid" or "early".

The example harvesting map shown in Figure 14 shows the distribution of crop zones A, B and C across the fields of Shelford potato tubers. The harvesting map was generated based on data indicating the stage at which senescence occurs. The harvesting map may be supplied to a crop harvesting apparatus, or the operator of a crop harvesting apparatus, to enable crops to be separated accordingly. Each of the samples were then harvested and their dry matter content was determined. Figure 15 shows a table containing the dry matter content for crop zones A1 -A3 ("Shelford-early), B1 -B3 ("Shelford-mid") and C1 -C3 ("Shelford-late"). As shown in this table, the dry matter content of the potato tubers grown in crop zone C, ie crops thai are late to senescence, is significantly lower than the dry matter content of the potato tubers grown in crop zones A or B, ie those that are earlier to senescence. It has been found that the lower the dry matter content of the potato crop, the higher the oil content of the potato crop when fried as crisps. It is expected that the oil content of the potato tubers grown in crop zone C would be 13% higher than the oil content of the potato tubers grown in crop zone A. It has also been found that the dry matter content of potato tubers does not change during storage. The above information therefore allows the potato tubers to be separated as to which crop zones have the appropriate level of dry matter content for specific uses of the potato tubers. Figure 16 shows a multispectral image of a portion of a group of crop fields, with harvesting areas and graded crop zones indicated. In particular, the crop zones are indicated by the grade that has been assigned to the crop zone which, in this example, are grades AA, AB and BA. The multi-spectral camera in this example has five bands, which are as follows:

Band Name Centre Wavelength (nm) Bandwidth FWH (nm)

Blue 475 20

Green 560 20

Red 668 10

Near IR 840 40

Red Edge 717 10

The filter transmissivity of the camera, as a function of wavelength, is shown in Figure 17. In addition, Figure 17 shows a typical plant reflectance as a function of wavelength, encompassing all five of these bands. Alihough the crop growth parameter discussed above is the Normalized Difference Vegetation Index (NDVI), there are a number of other crop reflectance indices that may also be suitable. These crop reflectance indices are described below. OSAVI (Optimised Soil-Adjusted Vegetation Index) - this index is a simplified version of SAVI (Soil-Adjusted Vegetation Index maps) to minimize the influence of soil brightness. This index is recommended to analyse crops in early to mid- growth stages variability in canopy density and is not sensitive to changing soil brightness as NDVI. This index is best used in areas with relatively sparse vegetation where soil is visible through the canopy. An example of an image providing data using this index is shown in Figure 18(a).

NDRE (Normalised Difference Red Edge) - this index is sensitive to chlorophyll content in leaves, variability in leaf area, and soil background effects. NDRE can therefore be used to analyse whether images obtained from multi-spectral image sensors contain healthy vegetation or not. It is similar to the Normalized Difference Vegetation Index (NDVI), but uses the ratio of near-infrared and the edge of red as follows:

The red edge is the term used to describe the part of the spectrum centred around 715 nm. NDRE uses a red edge filter to view the reflectance from the canopy of the crop. The red edge is a region in the red-NIR transition zone of vegetation reflectance spectrum and marks the boundary between absorption by chlorophyll in the red visible region, and scattering due to leaf internal structure in the NIR region. This allows the determination of many different variables for crop management. Understanding the levels of chlorophyll can provide the ability to monitor photosynthesis activity. An example of an image providing data using this index is shown in Figure 18(b). NDVI (Normalised Difference Vegetation Index) - this index contrasts the red and near-infrared bands of light reflected from plant leaves. It is a general indicator of canopy density and is frequently used to distinguish live green vegetation from soil. An example of an image providing data using this index is shown in Figure 18(c).

DSM (Digital Surface Model) - this model represents the Mean Sea Level elevations of the reflective surfaces of trees, buildings, and other features elevated above the "Bare Earth". An example of an image providing data using this index is shown in Figure 18(d).

DEM (Digital Elevation Model) - this model is a type of raster GIS layer. In a DEM, each cell of raster GIS layer has a value corresponding to its elevation (z-values at regularly spaced intervals). DEM data files contain the elevation of the terrain over a specified area, usually at a fixed grid interval over the "Bare Earth". The intervals between each of the grid points will be referenced to some geographical coordinate system (latitude and longitude or UTM (Universal Transverse Mercator) coordinate systems (Easting and Northing). For more detailed information in DEM data file, it is necessary that grid points are closer together. The details of the peaks and valleys in the terrain will be better modelled with small grid spacing than when the grid intervals are very large. DTM (Digital Terrain Model) - this model can be described as a three - dimensional representation of a terrain surface consisting of X, Y, Z coordinates stored in digital form. It includes not only heights and elevations but other geographical elements and natural features such as rivers, ridge lines, etc. A DTM is effectively a DEM that has been augmented by elements such as break lines and observations other than the original data to correct for artifacts produced by using only the original data. With the increasing use of computers in engineering and the development of fast three-dimensional computer graphics the DTM is becoming a powerful tool for a great number of applications in the earth and the engineering sciences.

NIR Reflectance - this is a near infrared intensity render. Vegetation will appear brighter, while non-vegetation will appear darker. CIR (Colour Infrared Composite) - this index combines the NIR, Red, and Green bands. Healthy vegetation reflects a high level of NIR and appears red. Dormant vegetation is often green or tan, while sandy soils appear light tan and clay soils dark tan or bluish green.

RGB Composite - this index is the natural colour composite using red, green, and blue bands. Similar in appearance to a satellite or aerial image or an image from a standard camera. DSM, DEM and DTM's can be produced using imaging cameras, lasers, such as a LIDAR system (Light Detection And Ranging) and radar, such as an SAR system (Synthetic-Aperture Radar).

In respect of imaging cameras, in addition to the multi-spectral camera described above, the imaging camera could comprise a greater number of bands, eg greater than 100 bands, such as a so-called hyper-spectral camera. Alternatively, the imaging device may comprise a system of multispectral cameras or hyper-spectral cameras. In relation to laser systems, 'Lidar' (also called LIDAR, LiDAR, and LADAR) is a surveying method that measures distance to a target by illuminating that target with a pulsed laser light, and measuring the reflected pulses with a sensor.

Differences in laser return times and wavelengths can then be used to make digital 3D-representations of the target. The name lidar, sometimes considered an acronym of Light Detection And Ranging (sometimes Light Imaging, Detection, And Ranging), was originally a portmanteau of light and radar.

In relation to radar systems, 'synthetic-aperture radar' (SAR) is a form of radar that is used to create two- or three-dimensional images of objects, such as landscapes. SAR uses the motion of the radar antenna over a target region to provide finer spatial resolution than conventional beam-scanning radars. SAR is typically mounted on a moving platform, such as an aircraft or spacecraft, and has its origins in an advanced form of side-looking airborne radar (SLAR). The distance the SAR device travels over a target in the time taken for the radar pulses to return to the antenna creates the large "synthetic" antenna aperture (the "size" of the antenna). Typically, the larger the aperture, the higher the image resolution will be, regardless of whether the aperture is physical (a large antenna) or "synthetic" (a moving antenna) - this allows SAR to create high-resolution images with comparatively small physical antennas.

To create a SAR image, successive pulses of radio waves are transmitted to "illuminate" a target scene, and the echo of each pulse is received and recorded. The pulses are transmitted and the echoes received using a single beam-forming antenna, with wavelengths of a meter down to several millimetres. As the SAR device on board the aircraft or spacecraft moves, the antenna location relative to the target changes with time. Signal processing of the successive recorded radar echoes allows the combining of the recordings from these multiple antenna positions - this process forms the "synthetic antenna aperture" and allows the creation of higher-resolution images than would otherwise be possible with a given physical antenna.

Airborne systems provide resolutions of about 10 cm, ultra-wideband systems provide resolutions of a few millimetres, and experimental terahertz SAR has provided sub-millimetre resolution in the laboratory.

Claims

1 . A method of agricultural harvesting, which method comprises the steps of:

(a) obtaining one or more images of a crop region,

2. A method as claimed in Claim 1 , wherein the method comprises either (i) measuring a crop growth profile that is a series of crop growth parameters measured at different times for each crop zone, or (ii) the crop growth parameter is a time, or an average of the times, at which one or more plants in each crop zone reach a particular stage of growth or retreat, and wherein the grade for each crop zone is determined by the crop growth parameter or the crop growth profile for that crop zone.

3. A method as claimed in Claim 1 or Claim 2, wherein the one or more images are generated from at least some visible wavelength light, eg red light, and/or at least some non-visib!e wavelength light, eg near-infrared light.

4. A method as claimed in any preceding claim, wherein the one or more images are obtained by a multispectral image sensor that detects wavelengths of at least 600-800nm.

5. A method as claimed in any preceding claim, wherein the one or more images are obtained by an imaging device mounted on a remotely piloted aircraft system.

6. A method as claimed in any preceding claim, wherein the crop zones comprise a plurality of zones within an area for harvest, the area for harvest corresponding to an area that would be harvested in a single operation, eg by a harvesting machine.

7. A method as claimed in Claim 6, wherein the area for harvest comprises one or more rows of plants, eg at least 2, at least 3, or at least 4 rows of plants.

8. A method as claimed in Claim 6 or Claim 7, wherein the crop zones comprise an array of zones within each area for harvest.

9. A method as claimed in any preceding claim, wherein the crop growth parameter may correspond to a count of the number of plants within each crop zone.

1 0. A method as claimed in any preceding claim, wherein the crop growth parameter may correspond to the area covered by the one or more plants in each crop zone,

1 1 . A method as claimed in any preceding claim, wherein the crop growth parameter may correspond to the Normalised Difference Vegetation Index (NDVI) of the crop zone, where the NDVI is calculated to be equal or proportional to (NIR - VIS) / (NIR + VIS), where VIS and NIR are the spectral reflectance

measurements acquired in the visible and near-infrared regions, respectively.

1 2. A method as claimed in any preceding claim, wherein the crop growth parameter may correspond to a time, or an average of the times, at which the one or more plants in each crop zone reach a particular stage of growth or retreat.

1 3. A method as claimed in any preceding claim, wherein the measured crop growth parameter for each crop zone takes the form of a crop growth profile, which is a series of crop growth parameters measured at different times, and the plurality of pre-deiermined grades each have a representative crop growth profile, against which the measured crop growth profiles are fit, with the grade for each crop zone determined by the representative crop growth profile against which the measured crop growth profile best fits.

14. A method as claimed in any preceding claim, wherein the pre-determined grades are indicative of the quality of the harvested crop.

15. A method as claimed in Claim 14, wherein the quality of the harvested crop includes any of storage stability, physical composition, chemical composition, appearance, and any other characteristic that is desired of the harvested crop.

16. A method as claimed in any preceding claim, wherein the crop is harvested by a crop harvesting machine, which comprises a vehicle, and a mechanism for removing the crop from the ground or from the plants.

17. A method as claimed in Claim 16, wherein the crop harvesting machine includes a plurality of containers, each container being for receiving a particular grade, or a particular group or range of grades, of the crop.

18. A method as claimed in Claim 16 or Claim 17, wherein the crop harvesting machine includes a system for identifying the location of the crop being harvested.

19. A method as claimed in any one of Claims 16 to 18, wherein the crop harvesting machine is operated to harvest crops from each crop zone in turn, one after another, with a sorting mechanism causing the crop from each crop zone to be transferred to an appropriate container, according to grade.

20. A method as claimed in Claim 19, wherein this operation is computer assisted or computer controlled, and utilises one or more data files containing data regarding the location of the crop zones and the pre-determined grade that has been assigned to each crop zone.

21 . A method as claimed in any preceding claim, wherein the crops are transferred, once harvested, into long-term storage, where the separated, graded crops are stored separately, and are treated differently, with respect to physical or chemical treatment.

22. A method as claimed in Claim 21 , wherein the separated, graded crops have different storage temperatures, humidity, atmosphere, and/or applied chemicals.

23. A crop harvesting apparatus comprising an imaging device for obtaining one or more images of a crop region, a data processor for assigning a plurality of crop zones to the one or more images of the crop region, measuring from the one or more images a parameter representative of crop growth for each crop zone, and assigning one of a plurality of pre-determined grades to each crop zone, the grade for each crop zone being determined by the growth parameter for that crop zone, and a crop harvesting machine for harvesting the crop by separating and at least temporarily storing crops according to the grades assigned to the crop zones.

24. A crop harvesting apparatus as claimed in Claim 23, wherein the apparatus comprises a data processor for measuring from the one or more images a parameter representative of crop growth for each crop zone, and either (i) measuring a crop growth profile that is a series of crop growth parameters measured at different times for each crop zone, or (ii) the crop growth parameter is a time, or an average of the times, at which one or more plants in each crop zone reach a particular stage of growth or retreat, and wherein the grade for each crop zone is determined by the crop growth parameter or the crop growth profile for that crop zone.

25. A crop harvesting apparatus as claimed in Claim 23 or Claim 24, wherein the crop harvesting machine is configured to receive a data file comprising data representing the location of a plurality of crop zones, and data identifying the predetermined grade assigned to each crop zone, and the crop harvesting machine is operable to harvest crops from each crop zone in turn, one after another, with a sorting mechanism causing the crop from each crop zone to be transferred to an appropriate container, according to grade.