Capitan - Flight Plan Generation
We are constantly working to reduce our turnaround time for Insights and therefore we have made the following improvements:
- Validate Obstacle - drone pilots will now be able to differentiate between the field's boundary obstacles and the pilot's manual obstacles. The pilot manual obstacles can change from flight to flight. This new feature provides a quick and easy way for pilots to reset the obstacles to the default and adapt the flight plan to the current situation in the field. Consequently, we are able to minimize even more the time spent by the pilot in the field and the turnaround time for the Taranis Insights.
Generate Flight Plan - we are enabling the pilot to view the image capturing waypoints in the drone flight plan already as part of the flight plan generation. This was previously not the case which did cause delays. Having the ability to view those waypoints at the flight plan generation stage will enable the pilot to troubleshoot and fix issues much earlier.
Capitan - Flight Missions Sorting Capabilities
Helping the pilots to navigate between his/her missions we have provided sorting capabilities by flight window, action, order ID, name, crop, and schedule date. By default, we sorted the missions by scheduled date but in many cases, the pilot might want to change the sorting in order to better plan his/her time in the field further reducing our Insights turnaround time.
Capitan - Collect more data of the images
In order to further improve the quality of the images, we now collect more data of the images taken such as: ISO, Aperture and Shutter Speed. This data will help us to set the right parameters to ensure high quality imagery regardless of lighting conditions, weather and time of day.
We have improved the performance of one of our major pilot operations tools to enable our drone pilots to quickly and efficiently complete their tasks before they head out to the field. In this case, the initial loading time was improved significantly - 10% of the time it took previously.
Taranis Artificial Intelligence: Improved Model for Insects Detection
In order to further increase the consistency and speed of our SmartScout Insights, we have made improvements to our AI capabilities for insect detection. These improvements will enable us to more quickly and accurately detect a wide variety of insects in corn, further improving the turnaround time for our Smart Scout Insights.
In order to increase the speed with which we are progressing towards full AI tagging of all of our Insights, we have created a separate simulation environment that enables us to test different AI algorithms without impacting the service / product that our customers are enjoying.
In the cycles management screen for multiple fields, row spacing was adapted to support localization - now showing the units according to metric/imperial selection, and the same for the single-cycle editing.
Insights present images with findings more accurately
Insights include a field map with dots that indicate the locations of all captured images.
White dots indicate images that have no findings
Red dots indicate images that have findings of the phenomena indicated in the insight. When opening an image with a red dot, you will see the findings marked with squares.
In some cases, insights presented white dots on the field map instead of red dots.
The fix now presents all images indications accurately - red and white dots.
New Field health insights are LIVE
With Field Health insight you can save time by identifying problem areas in the field more easily and reacting to crop stress before yield can be impacted.
The new version of Field Health provides you with "hotspot detection", indicating areas that are behaving significantly different than the rest of the field.
Together with SmartScout you will be able to “ground-truth” Field Health’s indication and see, on a leaf level, the complete and accurate cause for the stress.
The new Field Health insight not only marks hotspot areas that behave significantly different than the rest of the field but also differentiates areas that are below or above field average, which may be a result of different agronomic phenomena.
Field health - insight frequency creation
We added a mechanism to ensure that field health insights will be created with the frequency that our clients expect it to be, to provide the best service possible. In cases where weather conditions or satellite imagery quality don’t allow the insight to be created, it will be created as soon as clear and high-quality images are available.
Raizen - Weeds Coverage Product Pilot
We have successfully released a Proof Of Concept with our client in Brazil, aiming to validate the value in quantifying weeds pressure as the percentage of the area in the images, the identified weeds are covering. In order to do that, we’ve changed the tagging to be based on a grid of squares. In the image below - per grid square, if a weed exists - the square is tagged (see the green dot)
Then, we calculate the scores and the heatmap, based on this new way of tagging.
For this client, the Myfields CSV also shows 2 new columns, holding the new calculation score.
Another piece of functionality for the pilot is shown in the image below, a new type of Weeds insight report :
As part of the pilot, we will collect feedback and learnings, so we can plan our next steps with this new product solution, and what it takes in order to offer it in other regions as well.
A Sneak Preview - coming soon
Integration with Climate FieldView - field selection
We are enabling the control the user has over data transfer for specific fields, to Climate FieldView. The user will be able to control which fields to transfer the heatmap data, to any Climate FieldView user, by choosing specific fields that he can access to sync with the Climate account. This will prevent any potential data privacy issues between clients.
The user (retailer or grower) will select the fields to sync, and at any time this list of fields may be updated
We are changing the way we integrate data with Climate FieldView. Currently, upon connecting to VieldView, all the fields’ data under the organization are transferred to FieldView to be viewed. With the new integration flow, once a user attempts to connect, a list of all fields they have access to will open and the user will be able to select the fields to transfer the heatmap data for.
This will prevent any potential data privacy issues between clients under the same organization. At any point, the user will be able to update the list of fields to transfer data to Climate FieldView.
Nothing Detected Insight
We are aware of customer complaints about the fact that when we come to the field but have not found any problem, we do not present this information to customers, likely to think that we didn’t visit the field at all. We are also aware of the fact that this information is critical to know, and it’s a great and powerful statement to say that the field has no issue, which means that the yield probably will be higher than other fields.
From now and on, we will create an insight even if we didn’t find anything. This insight will be an insight that you are familiar with today, but the value will present the fact that we didn’t find anything. These insights will be available in the web app and CONNECT as well, and will behave as normal insights, including existence in other sections that we have such as “remote scouting” and “Map Layers”.
With this addition, you’ll also be able to point out the images themselves and verify that the field has no issues.