AUTOMATED SYSTEM FOR ANALYZING PLANT VIGOR

Abstract

A system for evaluating the quality and/or vigor of plants is described. Plants are planted in row sections and a cart is used to pass a radiometric sensor over the row sections. The cart has a radiometric sensor assembly positioned above the row section. Each sensor assembly generates a data signal and a computer receives and stores the data signals. The field cart is positioned above the row sections and measures the existence of plants in the row section and the quantity of vegetation in the row section.

Description

BACKGROUND OF THE INVENTION

The present invention relates to a system for automated system for analyzing plant vigor and/or quality. More specifically, the invention relates to a field cart or a tool bar mobile attachment for a vehicle for use in an automated system for measuring plant vigor and/or quality, and also to methods of selecting plants based on the automated system.

Agricultural small plot field research trials are designed to measure treatment effects. If plot quality is compromised, the intended treatment cannot be measured. Existing procedures require evaluation of plot quality approximately four to six weeks after planting. Plants are counted or visually scored for completeness of plant stand and vigor. Plots not meeting minimum quality standards are noted for exclusion from further analysis. Existing manual procedures are costly, labor intensive and not always precise.

SUMMARY OF THE INVENTION

The invention consists of a system and a field cart used in plant breeding programs to automate evaluating the quality and/or vigor or quality of plants, including specifically the evaluation of the quality and/or vigor of plants in a plant breeding program.

In a preferred embodiment, the invention consists of a system for automating the process of quantifying early season plot quality in agricultural small plot field research trials. The system uses active radiometric sensors to measure canopy spectral reflectance per row expressed in NDVI units (normalized difference vegetation index). Vegetation readings are separated from soil readings and reported as percent vegetation coverage and average NDVI.

The system automates the process of screening thousands of experimental plants for plant quality or vigor. Typically, evaluating plant vigor is a manual process that relies on several experienced technicians to make and record hundreds of evaluations per hour. This manual system uses a numerical rating system from one to nine, where one equals optimum vigor and nine equals plant death. Thousands of plots must be manually evaluated on a daily basis by multiple technicians. The evaluations are subjective because of differing biases and amount of experience of each technician. A technician can typically evaluate between 500 and 1000 plants per hour.

In one embodiment, the invention provides a system for evaluating the vigor of growing plants. The invention is used to evaluate the quality and/or vigor of plants growing in rows of a field. Apparatus for taking data regarding the quality of the plants in the rows are mounted on a field cart for easy transport in the field. The field cart includes a body supported on wheels above the plant canopy. A radiometric sensor is mounted on the body of the cart and positioned so that it looks down on a row as the cart is moved through the field. The number of sensors corresponds to the number of rows of soybean plants that are spanned by the cart so that each sensor is positioned above a row. As the cart is pushed down the plurality of rows, each sensor assembly collects data from the plants in the corresponding row and generates a data signal that is received and stored in a computer also mounted on the cart. Preferably, the position of each plant in each row of the field or range was recorded by GPS apparatus associated with a planter that planted the row and the field cart also includes GPS apparatus such that the data generated can be correlated with the recorded planting position and hence the identity of the seed planted at the location for use in a breeding program for developing improved varieties of plants.

In another embodiment, the invention provides a method of evaluating the quality and/or vigor of growing plants. The method includes planting a plurality of rowed plots in a field and positioning a wheeled field cart above the growing plants. The field cart includes a body with at least one sensor secured to the body. Each sensor generates a data signal and a computer receives and stores the data signals. The method also includes the steps of positioning each sensor above a single row of plants, scanning each plant in each row, transmitting a data signal from each sensor to the computer, and storing the data signals in the computer.

Other aspects of the invention will become apparent by consideration of the detailed description and accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a perspective view of a corn field.

FIG. 2 is a perspective view of a wheeled cart supporting apparatus included in the present invention shown spanning two rows of corn in a field.

FIGS. 3-5 are additional perspective views of the cart.

FIG. 6 is another perspective view of the cart shown spanning two rows of corn in a field.

FIG. 7 is a photograph of a sensor used in a system of the present invention.

FIG. 8 is a photograph of a section of a row in a field showing a single corn plant and a graph of the data taken by a system of the present invention over the row section.

FIG. 9 is a photograph of a section of a row in a field showing a number of corn plants and a graph of the data taken by a system of the present invention over the row section.

FIG. 10 is a photograph of a section of a row in a field showing a number of corn plants and a graph of the data taken by a system of the present invention over the row section.

FIG. 11 is a photograph of a section of a row in a field showing a number of corn plants and a graph of the data taken by a system of the present invention over the row section.

FIGS. 12 a-d are photographs of a section of a row in a field showing a number of soybean plants and a graph of the data taken by a system of the present invention over the row section.

FIG. 13 is a schematic diagram of some of the electrical components of a system of the present invention.

FIGS. 14 a-c are photographs of a tractor-mounted embodiment of the present invention.

FIGS. 15-17 are graphs of data collected by the field cart over a field plot with increasing sections of the crops removed from the rows and at three different soil threshold settings.

FIG. 18 is a graph of data collected by the tractor-mounted system over a field plot with increasing sections of the crops removed from the rows.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Before any embodiments of the invention are explained in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways.

The apparatus and methodologies described herein may make advantageous use of the Global Positioning Satellite (GPS) system to determine and record the positions of fields, plots within the fields and plants within the plots and to correlate collected plant condition data. Although the various methods and apparatus will be described with particular reference to GPS satellites, it should be appreciated that the teachings are equally applicable to systems which utilize pseudolites or a combination of satellites and pseudolites. Pseudolites are ground- or near ground-based transmitters which broadcast a pseudorandom (PRN) code (similar to a GPS signal) modulated on an L-band (or other frequency) carrier signal, generally synchronized with GPS time. Each transmitter may be assigned a unique PRN code so as to permit identification by a remote receiver. The term “satellite”, as used herein, is intended to include pseudolites or equivalents of pseudolites, and the term GPS signals, as used herein, is intended to include GPS-like signals from pseudolites or equivalents of pseudolites.

It should be further appreciated that the methods and apparatus of the present invention are equally applicable for use with the GLONASS and other satellite-based positioning systems. The GLONASS system differs from the GPS system in that the emissions from different satellites are differentiated from one another by utilizing slightly different carrier frequencies, rather than utilizing different pseudorandom codes. As used herein and in the claims which follow, the term GPS should be read as indicating the United States Global Positioning System as well as the GLONASS system and other satellite- and/or pseudolite-based positioning systems.

FIG. 1 illustrates an agricultural field 25 which has been planted in accordance with the methods described herein. A planter equipped with a high-precision GPS receiver results in the development of a digital map of the agricultural field 25. The map defined through this operation may become the base map and/or may become a control feature for a machine guidance and/or control system to be discussed in further detail below. The map should be of sufficient resolution so that the precise location of a vehicle within the area defined by the map can be determined to a few inches with reference to the map. Currently available GPS receivers, for example as the ProPak®-V3 produced by NovAtel Inc. (Calgary, Alberta, Canada) are capable of such operations.

For the operation, a tractor or other vehicle is used to tow a planter across the field 25. The planter is fitted with a GPS receiver which receives transmissions from GPS satellites and a reference station. Also on-board the planter is a monitoring apparatus which records the position of seeds as they are planted by the planter. In other words, using precise positioning information provided by the GPS receiver and an input provided by the planter, the monitoring apparatus records the location at which each seed is deposited by the planter in the field 25.

As the tractor and planter proceeds across field 25 to plant various rows of seeds or crops, a digital map is established wherein the location of each seed planted in field 25 is stored. Such a map or other data structure which provides similar information may be produced on-the-fly as planting operations are taking place. Alternatively, the map may make use of a previously developed map (e.g., one or more maps produced from earlier planting operations, etc.). In such a case, the previously stored map may be updated to reflect the position of the newly planted seeds. Indeed, in one embodiment a previously stored map is used to determine the proper location for the planting of the seeds/crops.

In such an embodiment, relevant information stored in a database, for example the location of irrigation systems and/or the previous planting locations of other crops, may be used to determine the location at which the new crops/seeds should be planted. This information is provided to the planter (e.g., in the form of radio telemetry data, stored data, etc.) and is used to control the seeding operation. As the planter (e.g., using a conventional general purpose programmable microprocessor executing suitable software or a dedicated system located thereon) recognizes that a planting point is reached (e.g., as the planter passes over a position in field 10 where it has been determined that a seed should be planted), an onboard control system activates a seed planting mechanism to deposit the seed. The determination as to when to make this planting is made according to a comparison of the planter's present position as provided by the GPS receiver and the seeding information from the database. For example, the planting information may accessible through an index which is determined according to the planter's current position (i.e., a position-dependent data structure). Thus, given the planter's current location, a look-up table or other data structure can be accessed to determine whether a seed should be planted or not.

In cases where the seeding operation is used to establish the digital map, the seeding data need not be recorded locally at the planter. Instead, the data may be transmitted from the planter to some remote recording facility (e.g., a crop research station facility or other central or remote workstation location) at which the data may be recorded on suitable media. The overall goal, at the end of the seeding operation, is to have a digital map which includes the precise position (e.g., to within a few inches) of the location of each seed planted. As indicated, mapping with the GPS technology is one means of obtaining the desired degree of accuracy.

As shown in FIGS. 1 and 2, the field 25 is planted with a plurality of rows 21 of plants, which in this embodiment are corn plants.

Different varieties of plants may be planted in the field 25 as part of a breeding program to evaluate the quality and/or vigor of the different varieties to the growing conditions in the field. For example, the field 25 may have soil of a high pH and the breeding program may be set up to determine varieties that are resistant, tolerant, or susceptible to soil with a high pH. Of course resistance, tolerance or susceptibility to a wide range of conditions can be evaluated and will be readily apparent to one of skill in the art.

The apparatus and methodologies described herein utilize radiometric crop sensor assemblies 40 that measure the reflectance and absorbance of one or more frequencies of light by plant tissues. There are two types of radiometric sensor assemblies 40, active sensor assemblies which use one or more internal light sources to illuminate the plants being evaluated, and passive sensor assemblies which use ambient light only. One suitable index in assessing crop conditions is the normalized difference vegetative index (NDVI). The NDVI was developed during early use of satellites to detect living plants remotely from outer space. The index is defined as NDVI=(NIR−R)/(NIR+R) where NIR is the reflectance in the near infrared range and R is the reflectance in the red range but other visual frequencies can be substituted for red. Preferred sensors for use with the present invention generate an output that is in NDVI units.

As shown in FIGS. 2-6, a preferred sensor assembly 40 is the GreenSeeker® RT100 sold by NTech Industries (Ukiah, Calif.), now a part of Trimble Navigation Limited (Sunnyvale, Calif.). In other embodiments, passive sensor assemblies that utilize ambient light are used.

As shown in FIG. 7, a radiometric sensor assembly 40 includes a casing 45, a light source 50 mounted in the casing 45, and a sensor 55 mounted in the casing 45. In some embodiments, the sensor assembly 40 includes a sensor module including the light source 50 and the sensor 55 and a control box electrically connected to the sensor module. In other embodiments, the sensor assembly 40 includes multiple sensors 55 and multiple light sources 50. As explained above, the sensor 55 is configured to measure the reflectance and absorbance of one or more frequencies of light by plant tissues and generate an output in NDVI units.

As shown in FIGS. 2-6, a field cart 60 includes a body 65, a pair of sensors 40 mounted to the body 65, a computer 75, and a power supply 80. The body 65 includes a substantially rectangular frame 85 supporting a workspace 90. The body 65 also includes four legs 95, each of the legs 95 extending substantially perpendicularly from the frame 85. A wheel 100 is mounted to each leg 95 opposite from the frame 85. The four wheels 100 are grouped as two front wheels and two rear wheels and as two right-side wheels and two left-side wheels.

Each sensor 40 is secured to the field cart 60. Each sensor 40 is electrically connected to the computer 75 and is powered by the power supply 80. In a preferred embodiment, the light source 50 transmits a narrow band of red and infrared light modulated at 50 ms. with the sensor assembly 40 configured to take twenty readings per second.

As shown in FIG. 13, the computer 75 includes a processor 145, a memory unit 150 electrically connected to the processor 145, a user interface 155 electrically connected to the processor 145, a display 160 electrically connected to the processor 145, and a GPS system 165 electrically connected to processor 145. The GPS system 165 can be a stand-alone component or physically integrated with the computer 75. The sensors assemblies 40 are electrically connected to the processor 145. The computer 75 is electrically connected to the power supply 80. A computer software program is used to calibrate, control and record data of the evaluation. The computer 75 is supported on the workspace 90. Alternatively, a GPS system 165 is not included in the computer 75.

The technician uses the field cart 60 to scan a section of two rows. During a scan, each sensor assembly 40 measures the reflectance and absorbance of one or more frequencies of light from a plant 20, if any, present in the row section in NDVI units. NDVI values are on a continuous numeric scale between zero and one, where a high number indicates a plant 20 with normal growth and a low number indicates a plant 20 that is adversely affected the growing conditions. The measured NDVI values are therefore represent the vigor of the plant 20 evaluated in the row section. The ability to make precise inferences is improved by using a continuous scale compared to the indexed numerical scale used with manual evaluation. The sensors 40 are calibrated to a known standard and provide consistent readings across an experimental field 25, thus reducing or eliminating the subjective variation across multiple technicians and the range-to-range, day-to-day variation of each technician.

In a preferred embodiment, the row sections are five feet long. Since the sensor 40 takes a reading every 50 ms, the number of readings or measurements taken in the five foot row section obviously depends on the speed the technician pushes the cart 60, but will typically be between 13 and 16 measurements. If the NDVI value for any of the measurement locations is above a selected value, the system will record a plant as present at that location, whereas if the NDVI value for any of the measurement locations is below the selected value, the system will record that no plant was detected. The selected value can be set by the technician and typically will be close to an NDVI reading of bare soil in the area of the field being measured, referred to herein as the soil threshold. The system generates a coverage ratio by taking the number plants detected and dividing by the number of measurement locations. The system, accordingly, yields a measurement of the number of plants that are growing in each row section, sometimes referred to as completeness of stand, and the average vigor of the plants in the row section, thus giving an objective evaluation of the quality of the stand of plants in each row section.

A field 25 is planted using multiple varieties of a selected crop according to a planned experiment, preferably using a planter that was equipped with a GPS device such that the location of each plant 20 is recorded together with the identity of the variety of seed planted in the corresponding location. The planting location and identity data is loaded into the computer 75. The computer program is used, together with the GPS system 165 to record the data gathered by the sensor assemblies 40 and associate that data with the location and identity data.

As shown in FIG. 2, a technician pushes the field cart 60 along the rows 21 such that a plurality of row sections of each of the two rows 21 pass between the left-side wheels 100 and the right-side wheels 100. The technician positions the field cart 60 so that the sensors 40 are positioned one each above a corresponding one of the two rows. The technician then triggers the sensor assemblies 40 to scan. The scan is triggered with the computer program by the user interface 155 and then the cart 60 is pushed by the technician down the two rows 21. Alternatively, the scan can be triggered using a switch, a button, or other known methods of generating an electrical signal. Approximately thirty-five hundred row sections can be screened in an hour using this automated method, more can be done with the mechanization of the movement of cart associated with the sensors. Compared to manual methods, the automated system is more objective and data collection at least two times faster.

When using a computer 75 without a GPS system 165, the technician manually verifies the location of field cart 60 at the start of a group of rows 21. When the field 25 is planted, a stake is placed in the ground at the start of a group of rows 21. A second stake is placed in the ground at a predetermined forward location. Each stake includes an individual identifier, for example a number or barcode. As the technician pushes the field cart 60 along the group of rows 21, the technician uses the stakes to verify the actual position of the field cart 60 compared to the expected location of the cart as determined by the computer program. For example, at the beginning of a group of rows 21, the computer program prompts the technician to verify the position of the field cart 60 using the stake at the beginning of the group of rows 21. Next, the technician inputs the identifier associated the stake and positions the field cart 60 above the first two row sections. Then, the technician triggers a scan. The computer program stores the data from the scan of the first two row sections and associates that data with the planned experiment. Then, the computer program automatically indexes to the next row sections as the cart is advanced by the technician. As the cart 60 nears the second stake, the computer program prompts the technician to verify the position of the field cart 60 using the second stake. In this manner, the position of the field cart 60 in the field 25 is tracked to ensure that the computer program is correctly associating the data collected by the sensor assemblies 40 with the preplanned experiment. The technician can use the computer program to monitor his position along the group of rows 21 relative to the stakes. If the field cart 60 is not in the expected position when the technician is prompted, the technician can use the computer 75 and computer program to correct the error or to identify the ranges 35 that were incorrectly associated with the planned experiment.

When using a computer 75 including a GPS system 165, the location of each row section of plants is automatically determined by the GPS system 165 and the data from the sensor assemblies 40 is automatically associated with planned experiment after a scan is performed. Alternatively, the GPS system 165 determines the location of each row section and the computer program automatically indexes to the next row section after a scan.

Alternatively, this field cart can be mobilized by addition of a motor, or it can be pulled behind a vehicle such as a truck, tractor, all wheel terrain vehicle, a mower, etc. An embodiment wherein eight sensors 40 are mounted on the toolbar of a tractorl 70 is illustrated in FIGS. 14a-c. Of course, the control components, including, for example, the computer 75 and GPS system 165 are also included in the tractor-mounted embodiment. The tractor-mounted embodiment, accordingly, is capable of taking measurements of eight rows simultaneously.

Example One

Evaluation of Early Stand Vigor of Corn Plants

An experiment was designed to evaluate the tolerance, resistance or susceptibility of a number of different varieties of corn to soil having a high pH. A field was planted with a plurality of rows of corn and each of the rows was divided into a plurality of row sections. Each of the row sections was planted with corn seed of a preselected one of the varieties.

Data were taken with the field cart when the plants had reached approximately the V3-V4 stage. FIGS. 8-11 show photographs of selected row sections and the data collected by the system from the depicted row section. In FIG. 8, the preselected value for determining the presence of a plant was set at 0.175, 14 measurements were taken and only a single measurement was above 0.175 so the coverage ratio of 14% was recorded. All 14 measurements of NDVI were averaged to generate a vegetative NDVI (Veg. NDVI), which in FIG. 8 was recorded as 0.209. In FIG. 9, 4 out of 15 measurement locations had an NDVI above 0.175 and so the coverage ratio 27% was recorded and the Veg. NDVI was 0.215. In FIG. 10, 12 out of 13 measurement locations had an NDVI above 0.175 and so the coverage ratio 92% was recorded and the Veg. NDVI was 0.22. In FIG. 11, 15 out of 15 measurement locations had an NDVI above 0.175 and so the coverage ratio 100% was recorded and the Veg. NDVI was 0.271.

Example Two

Evaluation of Early Stand Vigor of Soybean Plants

An experiment was designed to evaluate use of the system of the present invention to measure the defoliation of soybean plants. A field was planted with a plurality of rows of soybeans and each of the rows was divided into a plurality of row sections. Each of the row sections was planted with soybean seed of a preselected one of the varieties

In FIGS. 12a-d, soybean plants are growing in the row sections and the data taken from four row sections is shown. The preselected NDVI value for detecting the presence of a plant was set at 0.275. The first row section (FIG. 12a) had 16 out of 16 measurement locations detected with a plant present, giving a coverage ratio of 100% and reported an average NDVI of 0.86. The second row section (FIG. 12b) had 12 out of 13 measurement locations detected with a plant present, giving a coverage ratio of 92% and reported an average NDVI of 0.66. The third row section (FIG. 12c) had 12 out of 13 measurement locations detected with a plant present, giving a coverage ratio of 92% and reported an average NDVI of 0.55. The fourth row section (FIG. 12d) had 10 out of 14 measurement locations detected with a plant present, giving a coverage ratio of 71% and reported an average NDVI of 0.33. Note the identification by the system of bare soil between row sections.

Example Three

Measurement of Predetermined Gaps in Vegetation

A test field was planted with corn plants arranged in 8 sets of 10 blocks, each block having 16 rows. The length of each row was 17.5 feet. So as to evaluate the ability of the systems of the present invention to measure the effect of missing plants, gaps in the corn plants were created across the rows. Set 0 was the control and had no gap. Set 1 had a one foot gap created across the rows, Set 2 had a two foot gap, and so on, with Set 7 having a seven foot gap created across the rows. The field cart 60 and tractor-mounted embodiment 170 were both used to take measurements. Measurements were taken at different stages of plant growth; in this example, the sensors 40 were adjusted to be 32 inches above the plant canopy. There is a trade-off inherent in the speed of the cart 60 or tractor 170; if the speed is low, more measurements are taken in each row and so detection of gaps is improved but the time required increases, conversely, if the speed high, less time is consumed, but detection of gaps degrades. Measurements also were taken at varying soil thresholds to determine a satisfactory balance between over and under detection of the vegetation. To account for the differences between each sensor 40, the sensors 40 should be calibrated before data is taken.

The field was scanned using the cart 60 with the soil threshold set at zero. FIG. 15 shows the data taken at 2 mph. The “treatment” corresponds to the size of the gap in feet created across the rows and PVEGP is the percent vegetation. The letters immediately below the chart (A-E) reflect statistical differences between the treatments; if a treatment has a different letter then it is statistically different. In FIG. 16, the soil threshold was increased to 80% of the NDVI calibration for bare soil in the field in an effort to improve detection of the gaps. As can be seen from the data, each treatment is now statistically different from its adjoining treatment and the LSD (least squares difference) decreased from 4.1 to 3.3. The soil threshold was increased to 150% of the NDVI calibration for bare soil in the field and data taken at 1 mph is shown in FIG. 17. As can be seen, resolution of the gaps decreased. The tractor system 170 was used to take data at 1.5 mph and with the soil threshold set at 120% of the NDVI calibration for bare soil in the field. The data, shown in FIG. 18 gives plant vegetation percentage values that are close to theoretical.

Example 4

Several early season traits are useful to incorporate into planting breeding material and to quantify before plants are released commercially. One trait is emergence and the other is early seedling growth. Currently these traits are recorded by visual observation of a plant breeder or technician skilled in the art of making critical observations of plants and assigning a numeric value. A rating from 1 to 9 is used, 1 equals superior performance and 9 equals inferior performance. The values carry the individual biases and experience of the individual making the assessment.

By using the automated system for analyzing plant vigor unbiased, repeatable values can be assigned to plant material that quantify seedling emergence and early seedling growth. Readings taken within one to two weeks after initial emergence of individual plots of plants, each with different genetic makeup are separated for the ability to emerge uniformly and completely. Plots and the plants contained in the plots with superior emergence have uniformly high NDVI values, plots with inferior emerging plants have lower and more variable NDVI values.

Readings are taken about two to three weeks after the initial readings taken for emergence to quantify seedling growth. NDVI measurements are taken as in the initial readings and then a linear regression is used to measure the slope or rate of change from the initial readings to the second and later reading. Plots with a larger slope are considered superior in seedling growth. Plots with shallow slope are considered inferior in seedling growth. This trait is improved over current systems since the current system is a visual assessment of the condition among and between plots of plants at the time of assessment and not necessarily the change of each plot over time.

Claims

1. A system for evaluating the quality and/or vigor of growing plants, the system comprising: (a) a row of plants;(b) vegetative sensing apparatus;(c) a vehicle mounting the sensing apparatus for transport of the sensing apparatus over a section of the row of plants;(e) a data signal generated by the sensing apparatus corresponding to the evidence and quantity of the plant or plants in the row section; and(f) a computer for receiving and storing the data signal associated with each row section.

2. A system of claim 1, further comprising location information for each row section accessible by the computer and location determining apparatus on the vehicle to generate location information of the sensing apparatus to correlate the data signal to the corresponding row section.

3. A system of claim 2, wherein the location determining apparatus includes a global positioning satellite (GPS) system.

4. A system of claim 1, wherein the computer generates data corresponding to the completeness of stand of the plants in the row section.

5. A method for measuring the quality and/or vigor of growing plants, comprising the steps of: (a) planting seed of a selected variety of the plants in a row section and recording the position of the row section;(b) growing the plants to a selected stage for evaluation of quality and/or vigor;(c) collecting radiometric sensor data from each row section corresponding to the evidence and quantity of a plant or plants in the row section; and(d) analyzing the sensor data to generate a measure of the completeness of stand of plants in the row section.

6. The method of claim 5, wherein GPS is used to record the location of seed planted in the row sections.

7. The method of claim 6, wherein GPS is used to correlate the sensor data to the location of the row sections.

8. The method of claim 5, wherein varieties of plants of known completeness of stand are included as check plants.

9. The method of claim 5, wherein the sensor is mounted on a vehicle and supported above the plants.

10. A method of plant breeding, comprising the steps of: (a) planting seed of a selected variety of the plant in a row section and recording the position of the row section;(b) growing the plants to a selected stage for evaluation of quality and/or vigor;(c) collecting radiometric sensor data from each row section corresponding to the evidence and quantity of a plant or plants in the row section;(d) analyzing the sensor data to generate a measure of the quality and/or vigor of the variety of plant in the row section; and(e) using the measure of quality and/or vigor of the variety as a basis for selecting between plants in a plant breeding program.

11. A method for developing seeds bred for uniformity of seedling emergence, comprising the steps of: (a) planting seed of selected germplasm of the plants in a row section;(b) growing the plants to a selected stage for evaluation;(c) collecting automated sensor data readings from the plants at different times from emergence until said selected stage for evaluation to quantify seedling growth;(d) measuring the change between the initial automated data readings to the later in time data readings;(e) analyzing the automated sensor data readings to generate a measure of the uniformity of seedling emergence of the germplasm of plants(f) selecting plants with uniformity of seedling growth for further breeding(g) breeding a new variety with the uniformity of seedling growth; and(h) harvesting seed from the new variety or progeny evidencing the uniformity of seedling growth.