Description:FIELD OF THE INVENTION
This invention relates to Real-Time Morphological Feature Detection in Strawberries Using AI
BACKGROUND OF THE INVENTION
Manual monitoring and grading of strawberry morphology are a heavy toil; it is endless, acting. It is also quite subjective, leading to variations of quality grading and phenotyping. Traditional methods could not scale-up and contract to accurately capture subtle changes in plant traits occurring during time. Due to the advances in precision agriculture, developments in real-time accurate systems to monitor and analyse strawberry morphology during different growth stages are yet to be realized. Current solutions are predominantly image-data based or environmental context-based and simply do not provide the required yields in terms of accuracy. Open-field setups impose restrictions in terms of power and connectivity, further posing difficulties in deploying any form of intelligent monitoring system. This calls for a more robust, solar-powered kind of system able to combine sensor and image data information for cloud-based analyses so that it transcends these shortcomings, thereby ensuring accurate morphological assessment.
In the world of horticulture, strawberries are considered among the most profitable crops, with quality and visual appeal being major factors of consumer preference and market potential. Accurate morphometric evaluation of strawberries, in terms of size, shape, color, surface texture, and ripeness, among others, is essential for breeding, grading, and postharvest operations. Most of these features have nevertheless been assessed through manual and subjective procedures and are scarce in scalability. When inspections are conducted by humans, fatigue, judgment variability, or even environmental interference slide into the equation, all of which serve the eye-grader poorly in large-scale or real-time environment monitoring.
Though computer vision and AI have lately been booming in image-based crop phenotyping automation, most of the research and applications pertinent to strawberry monitoring have treated single aspects-cum-defect detection or fruit classification-based on static image data, while these models are usually trained in controlled environments and rarely generalize well in fields as variables like lighting, background, and occlusion affect the accuracy unfavorably. Some AI-based models use only visual data without considering multimodal inputs like environmental or sensor-based data (temperature, humidity, moisture), which adversely affects morphological variation-specific analysis at different developmental stages.
Another prominent gap in the literature is in the shortage of integrating energy-efficient, field-deployable systems. Most systems need high-end computing infrastructure with wired connectivity, which makes these systems impractical in distant agricultural locations. Little work has been done on decentralized edge-based AI systems that work on solar power and can transfer data to cloud environments using wireless networks. Further, temporal analysis of morphological traits, i.e., characterizing changes in traits over time, rarely finds mention in the current literature, though the very utility behind it is to enable ripening stage prediction, growth-abnormality detection, and optimization of harvest timing.
Besides this, the datasets available for training AI models in strawberry morphology are either proprietary or limited in their size and variation scope since staged annotations are missing across diverse cultivars and environmental conditions. This sets a barrier for the development of generalized models that could work on different field scenarios.
In short, it calls for an integrated, low-powered, real-time system that will combine sensor and image data, work autonomously in the open field environment, and assist AI with high-accuracy morphological analysis of strawberry at the different stages of growth. This research bridges that gap by designing a solar-powered Wi-Fi-enabled dual computing unit system that captures multimodal data and applies pretrained models for temporal morphological feature extraction of strawberries.
US1118151B2 The present invention relates to systems and methods for monitoring agricultural products. In particular, the present invention relates to monitoring fruit production, plant growth, and plant vitality. According to embodiments of the invention, a plant analysis system is configured determine a spectral signature of a plant based on spectral data, and plant color based on photographic data. The spectral signatures and plant color are associated with assembled point cloud data. Morphological data of the plant can be generated based on the assembled point cloud data. A record of the plant can be created that associates the plant with the spectral signature, plant color, spectral data, assembled point cloud data, and morphological data, and stored in a library.
RESEARCH GAP: This patent uses spectral and point cloud data for plant analysis, while the proposed system combines synchronized non-visual sensors and high-resolution imaging with AI for real-time, solar-powered morphological monitoring of strawberry plants.
CN111814622B The invention discloses a method, a system, equipment and a medium for identifying crop disease and insect pest types, which are characterized in that firstly, a training set and a verification set are obtained, an countermeasure network is constructed and generated, and the countermeasure network is generated through training of the training set and the verification set, so that a crop coding model is obtained; meanwhile, constructing a residual attention network, and obtaining a plant disease and insect pest classification model after training through a training set and a verification set; and aiming at the picture needing to be subjected to the disease and pest type identification, taking the picture as a test sample, firstly inputting the test sample into a crop coding model to determine whether the picture is a crop picture, if so, inputting the test sample into a disease and pest classification model, and identifying the disease and pest type in the picture through the disease and pest classification model. Based on the identification method provided by the invention, the crop disease and pest types can be accurately and rapidly detected.
RESEARCH GAP: This patent focuses on disease and pest identification using image-based deep learning models, while the proposed system detects morphological features and growth patterns of strawberry plants using a combination of non-visual sensors and synchronized imaging in a solar-powered field-deployable setup.
None of the prior art indicate above either alone or in combination with one another disclose what the present invention has disclosed. This invention relates to Real-Time Morphological Feature Detection in Strawberries Using AI
SUMMARY OF THE INVENTION
This summary is provided to introduce a selection of concepts, in a simplified format, that are further described in the detailed description of the invention.
This summary is neither intended to identify key or essential inventive concepts of the invention and nor is it intended for determining the scope of the invention.
To further clarify advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings.
An AI-powered system for real-time detection and analysis of morphological traits in strawberry plants is proposed to be fully integrated. This system works autonomously outdoors in an agricultural setting and runs on renewable sources of energy, which makes it suitable to be installed in remote sites without the usual electricity supply. The system design revolves around two independent yet interconnected subsystems: Sensor Computing Unit (Unit A) and Image Detection Computing Unit (Unit B). Each unit operates independently, with its own embedded computing device, real-time clock (RTC), solar panel, rechargeable battery setup, and wireless communication module-for collecting multimodal data set-environmental and visual-and transmitting it to a common cloud server for AI-based processing.
BRIEF DESCRIPTION OF THE DRAWINGS
The illustrated embodiments of the subject matter will be understood by reference to the drawings, wherein like parts are designated by like numerals throughout. The following description is intended only by way of example, and simply illustrates certain selected embodiments of devices, systems, and methods that are consistent with the subject matter as claimed herein, wherein:
FIGURE 1: OVERALL ARCHITECTURE
FIGURE 2 SENSORY DATA COLLECTOR DEVICE
FIGURE 3 EDGE VISION DATA COLLECTOR DEVICES
The figures depict embodiments of the present subject matter for the purposes of illustration only. A person skilled in the art will easily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the disclosure described herein.
DETAILED DESCRIPTION OF THE INVENTION
The detailed description of various exemplary embodiments of the disclosure is described herein with reference to the accompanying drawings. It should be noted that the embodiments are described herein in such details as to clearly communicate the disclosure. However, the amount of details provided herein is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the scope of the present disclosure as defined by the appended claims.
It is also to be understood that various arrangements may be devised that, although not explicitly described or shown herein, embody the principles of the present disclosure. Moreover, all statements herein reciting principles, aspects, and embodiments of the present disclosure, as well as specific examples, are intended to encompass equivalents thereof.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms “a",” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises,” “comprising,” “includes” and/or “including,” when used herein, specify the presence of stated features, integers, steps, operations, elements and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or groups thereof.
It should also be noted that in some alternative implementations, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may, in fact, be executed concurrently or may sometimes be executed in the reverse order, depending upon the functionality/acts involved.
In addition, the descriptions of "first", "second", “third”, and the like in the present invention are used for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Thus, features defining "first" and "second" may include at least one of the features, either explicitly or implicitly.
Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which example embodiments belong. It will be further understood that terms, e.g., those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
An AI-powered system for real-time detection and analysis of morphological traits in strawberry plants is proposed to be fully integrated. This system works autonomously outdoors in an agricultural setting and runs on renewable sources of energy, which makes it suitable to be installed in remote sites without the usual electricity supply. The system design revolves around two independent yet interconnected subsystems: Sensor Computing Unit (Unit A) and Image Detection Computing Unit (Unit B). Each unit operates independently, with its own embedded computing device, real-time clock (RTC), solar panel, rechargeable battery setup, and wireless communication module-for collecting multimodal data set-environmental and visual-and transmitting it to a common cloud server for AI-based processing.
The Sensor Computing Unit gathers environmental and contextual data from the strawberry growing environment. It comprises several sensors, including temperature and humidity sensors to detect climatic conditions, a soil moisture sensor to determine the water level near the roots, a soil pH sensor to check acidity and its effect on nutrient absorption, a light intensity sensor to monitor sunlight exposure, and a color sensor to complement fruit color calibration. Also included is a load cell (or weight sensor), which registers the weight of each fruit, and here the weight is directly related to the level of maturity. These sensors are controlled by a microcontroller-based computing unit, such as the ESP32 or Raspberry Pi Zero W, which receives the sensor data and then applies timestamps using an RTC module. The data undergoes wireless transmission through Wi-Fi to the cloud server for further processing at fixed intervals.
But the Image Detection Computing Unit compensates for the capturing of high-resolution images and visual data of the strawberry plants. It incorporates the HD RGB camera as the predominant device for image acquisition, capturing fine details to extract morphological characteristics such as shape, size, and surface peculiarities. Optionally, a line-scan camera can be integrated to serve conveyor or mobile continuous scanning, and a depth camera (e.g., Intel RealSense D435) can be used to build three-dimensional fruit profiles for accurate volume and contour measurement. In certain setups, an IR or NIR camera will be added for detecting ripeness and internal defects invisible to normal visible-spectrum imaging. This system is powered by an advanced edge computing device like Raspberry Pi 4 or NVIDIA Jetson Nano that performs light image processing and initial inference. An RTC module is also included, which guarantees synchronization with the sensor data. Like the Sensor Unit, it is powered by a solar battery system and wirelessly transmits image data to the cloud.
The two computing units aim for power efficiency and reliability. Being solar-powered, these units function uninterruptedly amid agricultural fields outdoors, while RTC modules maintain the time alignment and long-term data logging. A Wi-Fi connection is used between these units and the cloud server; however, in the absence of Wi-Fi, the system can be optionally fitted with GSM or LoRaWAN modules for maintaining connectivity. This conceptualization ensures that the system is not only field deployable but also scalable across various installations.
These servers host the cloud server from where the central processing vision synchronizes the data from sensors and images first. This includes synchronization based on timestamps, normalization of brightness and contrast of images, and calibration of environmental contrasts. The cleaned and structured data is then injected into a pretrained machine learning model that is capable of extracting in-depth morphological features. This AI model is generally deployed as a convolutional neural network (CNN) or a transformer-based vision model trained on a wide array of strawberry images and sensor data. It implies that the model conducts the identification of some important morphological-related traits such as fruit size (length, width, and area), morphotype classification (round, conical, irregular), surface texture analysis (bruises, mold, cuts), the distribution of seed patterns, calyx positioning, and color grading for ripening evaluation.
The result of the AI model makes a complete morphological profile of every strawberry sample, along with the growth patterns observed over time. That could be displayed either via a web dashboard or integrated into farm management systems for relevant decision-making. This supports applications such as automated phenotyping for breeding programs, yield prediction, quality control for harvesting, and diagnosis of developmental or environmental disorders. The merger of advanced sensing, edge computation, autonomous solar power, and cloud-based AI analytics creates an encompassing application of precision agriculture and smart horticulture, designed specifically these are what the challenges of strawberry production.
The algorithm employed in the system is a pre-trained machine learning model based on convolutional neural networks (CNNs), specifically for morphological feature extraction in strawberries. It ingests both the data i.e. HD image data and environmental data collected by the means of sensors from the field while synchronizing them with real-time timestamps for exact temporal alignment. The algorithm examines multimodal inputs for analysing and observing key characteristics like fruit colour uniformity, shape, surface defects and ripening stage. The model is trained on a huge labelled dataset of different types of strawberries and various conditions of growing; the model creates high precision and reliability. Through continuous updates and learning, it also adapts to the variability of the field. With a fusion of sensory and visual data, the algorithm also improves its accuracy of morphological analysis and grading. This allows for the provision of automation and decision support in strawberry quality monitoring and cultivation. The algorithm used is as follows:
Algorithm Used
BEGIN SYSTEM
// ---------- Initialization Phase ----------
INITIALIZE Sensor_Unit_A with:
TemperatureSensor
HumiditySensor
SoilMoistureSensor
LightSensor
PH_Sensor
ColorSensor
WeightSensor
RTC_A
WiFi_Module_A
Battery + Solar_Panel
INITIALIZE Image_Unit_B with:
RGB_Camera
Optional_Line_Scan_Camera
Optional_Depth_Camera
Optional_IR_Camera
RTC_B
WiFi_Module_B
Edge_Computer (Jetson or Pi)
Battery + Solar_Panel
INITIALIZE Cloud_Server with:
Data_Preprocessing_Module
ML_Model (Pretrained)
Database_Storage
Dashboard_Interface
// ---------- Data Collection Loop ----------
LOOP Every T_minutes:
// Sensor Unit A Data Collection
Sensor_Data ← {}
Sensor_Data.Timestamp ← RTC_A.GET_TIME()
Sensor_Data.Temperature ← TemperatureSensor.READ()
Sensor_Data.Humidity ← HumiditySensor.READ()
Sensor_Data.SoilMoisture ← SoilMoistureSensor.READ()
Sensor_Data.LightIntensity ← LightSensor.READ()
Sensor_Data.PH ← PH_Sensor.READ()
Sensor_Data.Color ← ColorSensor.READ()
Sensor_Data.Weight ← WeightSensor.READ()
// Transmit Sensor Data to Cloud
WiFi_Module_A.SEND_TO_CLOUD(Sensor_Data)
// Image Unit B Data Collection
Image_Data ← {}
Image_Data.Timestamp ← RTC_B.GET_TIME()
Image_Data.RGB_Image ← RGB_Camera.CAPTURE()
IF Line_Scan_Camera.ENABLED THEN
Image_Data.Line_Scan ← Line_Scan_Camera.CAPTURE()
IF Depth_Camera.ENABLED THEN
Image_Data.Depth_Map ← Depth_Camera.CAPTURE()
IF IR_Camera.ENABLED THEN
Image_Data.Thermal_Image ← IR_Camera.CAPTURE()
// Transmit Image Data to Cloud
WiFi_Module_B.SEND_TO_CLOUD(Image_Data)
END LOOP
// ---------- Cloud Server Processing ----------
ON Cloud_Receive(Sensor_Data, Image_Data):
// Synchronize & Preprocess
Synced_Data ← PREPROCESS(Sensor_Data, Image_Data)
// Extract Morphological Features
Morph_Features ← ML_Model.EXTRACT_FEATURES(Synced_Data)
// Save and Update Dashboard
Database.SAVE(Morph_Features)
Dashboard.UPDATE(Morph_Features)
END Cloud_Receive
// ---------- AI Inference Tasks ----------
FUNCTION EXTRACT_FEATURES(Synced_Data):
Image ← Synced_Data.RGB_Image
Sensor_Values ← Synced_Data.Sensor_Data
// AI Model predicts:
Size ← MODEL.Predict_Size(Image)
Shape ← MODEL.Predict_Shape(Image)
Color ← MODEL.Analyze_Color(Image, Sensor_Values.Color)
Ripeness ← MODEL.Grade_Ripeness(Color, Sensor_Values.LightIntensity)
Defects ← MODEL.Detect_Defects(Image)
Volume ← MODEL.Estimate_Volume(Synced_Data.Depth_Map)
RETURN {
Timestamp: Synced_Data.Timestamp,
Size: Size,
Shape: Shape,
Color: Color,
Ripeness: Ripeness,
Defects: Defects,
Volume: Volume
}
END FUNCTION
END SYSTEM
ADVANTAGES OF THE INVENTION:
The key advantages that make it a valuable tool for modern agriculture are:
• The proposed system monitors the strawberry plants using the collected image and sensor data. It collects the morphological features of the strawberry plant like shape, size, ripeness and surface defects in real-time. This enables in early detection of issues and supports timely by data-driven decisions on care and harvesting.
• Both of the computing units operates independently using rechargeable batteries and solar panels. Hence, making the system to work 24/7 in remote fields without any external power and power grids. The system supports sustainable agriculture and also reduces energy dependency.
• The system uses a combination of environmental sensor data as well as HD-image inputs. This multimodal approach helps in enhancing the reliability on extraction of morphological feature. It also ensures accurate detection under varying moisture, growth and light conditions.
• Each unit of the system is self-contained and wirelessly connected to the cloud server for data storage. This multiple units can be deployed across large farms without any complex wiring. This supports easy maintenance and cost-effective-expansion.
• A ML model which is pretrained by existing data analyzes the data to classify fruit quality and features. It provides consistent, objective grading for shape, ripeness, size and defects. This helps in yield forecasting and improves post-harvest value and sorting efficiency. 
, Claims:1. A solar-powered artificial intelligence system comprising two independent computing units, one configured for high-definition image capture and another for environmental sensor data collection, each unit including a real-time clock, a rechargeable solar-powered battery, and a Wi-Fi communication module, wherein the collected data is transmitted to a cloud server for preprocessing and feature analysis using a machine learning model.
2. The system as claimed in claim 1, wherein the sensor computing unit comprises a computing device connected to environmental sensors including humidity, soil moisture, temperature, pH, load cell, and light intensity sensors for collecting periodic data with real-time timestamps.
3. The system as claimed in claim 1, wherein the image detection unit comprises a high-definition camera and an edge processor configured for initial image handling, wherein each image is timestamped and wirelessly transmitted to the cloud server.
4. The system as claimed in claim 1, wherein the cloud server preprocesses the synchronized sensor and image data by aligning timestamps and normalizing inputs before feature extraction.
5. The system as claimed in claim 1, wherein a pretrained machine learning model processes the synchronized multimodal data to extract morphological features of strawberry plants.
6. The system as claimed in claim 1, wherein the extracted morphological features comprise fruit colour, shape, size, surface defects, and ripeness.
7. The system as claimed in claim 1, wherein the feature extraction enables automated grading and quality assessment of strawberries in real time.
8. The system as claimed in claim 1, wherein the synchronized multimodal analysis supports growth monitoring and developmental stage analysis of strawberries.
9. The system as claimed in claim 1, wherein the solar-powered rechargeable battery enables continuous operation of the system in open-field agricultural environments without external electricity.
10. The system as claimed in claim 1, wherein the wireless communication modules enable deployment of multiple units across large fields with centralized cloud-based processing and storage.