WO2022211644A1

WO2022211644A1 - A handheld device and method for cacao classification and maturity assessment

Info

Publication number: WO2022211644A1
Application number: PCT/PH2021/050011
Authority: WO
Inventors: Jan Joevil RAZON; Ryann ALIMUIN; Dunamis CRISOSTOMO; Paul Michael Edward ELIZES; Ryan Joshua FLORES; Christian Kurt HIQUIANA; Marivel NERY
Original assignee: Technological Institute Of The Philippines – Quezon City
Priority date: 2021-03-29
Filing date: 2021-05-15
Publication date: 2022-10-06

Abstract

The present invention discloses a handheld non-invasive dielectric spectroscopy device and the method thereof for determining the variety and maturity level of cacao pods. The device operates in two circuits: the variety classification circuit and the maturity level assessment circuit. In the variety classification circuit, a camera module is used to capture an image of the cacao pod, then it will be analyzed using a pre-trained neural network and will further be processed through a microprocessor. For the maturity assessment, a microwave signal is sent through the pod and will be analyzed by the differential amplifier. After which this signal is processed, analyzed and evaluated by the microprocessor to yield identify the maturity level. The variety and maturity level will then be displayed on an LCD screen for ease of understanding to the farmers.

Description

A HANDHELD DEVICE AND METHOD FOR CACAO CLASSIFICATION AND

MATURITY ASSESSMENT

TECHNICAL FIELD OF THE INVENTION

This invention relates generally to a non-invasive dielectric spectroscopy device that utilizes deep learning to extract information about agricultural crops and more specifically to a handheld non-invasive dielectric spectroscopy device utilizing a convolutional neural network to identify the variety classification and maturity level of cacao pods.

BACKGROUND OF THE INVENTION

The traditional method of harvesting and sorting of cacao pods relies heavily on the farmer’s manual assessment of the physical characteristics of the pod, such as color, or by flicking the pod. However, these methods are highly prone to human errors and normally result to misevaluation and premature harvesting of the pods.

For similar reasons, advancements in technology have been incorporated into the harvesting and sorting of various fruits and vegetables.

US 8,072,605 B2 (‘605) discloses a handheld non-destructive apparatus for determining the quality of fruit and vegetable products comprising a radiating light source, a detecting device for detecting a return radiation from the fruit and vegetable products, a sensor to receive part of the return radiation, a differentiating and amplifying device to provide the difference of absorbance of wavelengths, a central control and processing unit for processing, and a display for displaying the result of the processed signal. One drawback of ‘605 is that it only uses light within the 670-720 nm wavelength. Visible light can only reach until the epidermis of the cacao pod and hence cannot be effectively used to penetrate through the pod and obtain information on its maturity level.

Furthermore, ‘605 does not incorporate or suggest a method of determining the variety of a crop, an information that is very helpful in the sorting process after harvesting of the cacao pods. SUMMARY OF THE INVENTION

Accordingly, one object of the present invention is to provide a non-destructive method and device for identifying both the variety and the maturity level of cacao pods. Another object of this invention is to provide a dielectric spectroscopy device that is portable and can easily be used in the harvesting and sorting of cacao pods.

In the first aspect of this invention, there is provided a device for determining the variety classification and maturity level of cacao pods comprising a variety classification circuit and a maturity assessment circuit. The variety classification circuit comprises a camera module for capturing the image of the pod and a pretrained convolution neural network for processing the captured image, and a microprocessor for further processing. The maturity assessment circuit comprises a microwave signal generator, a resonator to act as a transceiver antenna, a detector for receiving the reflected signal, a differential amplifier for comparing voltages, and a microprocessor. In the second aspect of the invention, there is provided a method for determining the variety classification and maturity level of cacao pods comprising the steps of producing a microwave frequency through an oscillator, directing signal from said microwave frequency towards the pod through a resonator, penetrating through the pod with the said signal, receiving a reflected signal from the pod through the resonator, detecting voltage of the reflected signal through an RF detector, comparing the voltage from the RF detector with a reference DC voltage through a differential amplifier, sending output from the differential amplifier to the microprocessor, analyzing and evaluating said output, and displaying the results through a display screen.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows the left-side view of the present invention.

Figure 2 shows the right-side view of the present invention Figure 3 shows the back view of the present invention.

Figure 4 shows the front view of the present invention.

Figure 5A shows the bottom view of the antenna of the present invention.

Figure 5B shows the bottom view of an embodiment of the antenna of the present invention. Figure 6 shows the top view of the present invention.

Figure 7 shows the operation process of the variety classification circuit and maturity assessment circuit.

Figure 8 shows an embodiment of the oscillator schematic diagram.

Figure 9 shows an embodiment of the differential amplifier schematic diagram. Figure 10 shows an embodiment of the RF detector schematic diagram. DETAILED DESCRIPTION OF THE INVENTION

The details of the device and method of determining the variety and maturity level of cacao pods are as follows.

This handheld non-invasive dielectric spectroscopy device utilizes the technology of deep learning and microwave frequency signal spectroscopy. As shown in Figure 1, the spectroscopy device 1 is housed by a chassis and has a microwave frequency circuit (not shown) built inside an overhead case 2 which is connected by handle 3 to the battery chassis 4 containing the battery management system 6. The entire chassis of device 1 is made up of a 3D-printing filament to make the device portable.

The overhead case 2 is secured using overhead case screws 5a. The handle 3 is connected to the overhead case 2 and the lower side of the handle 3 using screws 5b and to the battery chassis 4 using screws 5c. The battery chassis 4 is secured using a battery chassis lock 7.

A multifunctional trigger 8 is located below the overhead case 2 and attached to the handle 3 to provide an easy access for the user when controlling the on and off state of the device.

An antenna 9 which focuses the signal being emitted by the open-ended coaxial cable (not shown) from the microwave frequency circuit is disposed adjacent the overhead case and oriented towards the targeted cacao pods to be analyzed.

An image capturing mount 10 for mounting a camera module is also oriented at the side facing the targeted cacao pods. As shown in Figure 2, all the parts in the left and right side of the device are symmetrically disposed.

Shown in Figure 3 is a thin-film-transistor liquid crystal display (TFT LCD) screen 11 disposed at the back of the device and facing the user. This enables the user to view the results of the variety classification and maturity level assessment.

Shown in detail in Figures 4, 5A and 6 are the camera module 10a mounted at the image capturing mount 10 and above the antenna 9 so that the user can easily target the cacao pods when operating the device 1.

In another embodiment, as shown in Figure 5B, the antenna 9 is a 31 -mm parabolic dish. Referring to Figure 7, the method of determining the variety and maturity level of the cacao pods is disclosed. The method has two schemes: the maturity level assessment and the variety classification.

The maturity assessment starts with the oscillator A producing the microwave frequency required. Acting as a transceiver antenna, the resonator B then directs the reflected signal towards the pod where the signal will penetrate through and the resonator B will then receive back the reflected signal. Since aiming at the cacao pods is to be done manually by the user, improperly positioning the resonator may occur and can lead to errors.

Once reflected, the RF detector C detects the voltages to be used by the differential amplifier D. The differential amplifier compares the two inputs, specifically, the detected voltage from the RF detector and the reference DC voltage E. The microprocessor F1, preferably a Raspberry Pi 4B ™, will analyze and evaluate the output from the differential amplifier D. The results will then be displayed through an LCD screen H to be viewed by the user.

In one embodiment of the present invention, the classification for the pod’s maturity level includes: immature, mature, and over-mature.

In another embodiment of the present invention, the classification is minimized into two: either mature, known as harvestable pods, or not mature, which is non- harvestable. This minimized classification is based on the close similarity in the characterization of the immature and the over-mature pods.

The variety classification starts with the camera module G capturing the image of the cacao pod. The image would then be analyzed by the microprocessor F with the aid of the pre-trained Convolutional Neural Network F2. The assessed variety of the cacao pod will also be displayed on the LCD screen H.

In one embodiment of the present invention, the variety classifier includes three (3) common types of cacao found in the Philippines: the BR-25, UF-18, and PBC 123. In another embodiment of the present invention, the LCD screen also displays the battery level and the cacao image, along with the maturity level and variety results.

Figure 8 shows an embodiment of the present invention wherein a series-tuned microwave frequency oscillator operating at 2.45GHz is used to serve as the frequency generator of the device, producing a fixed microwave frequency.

Figure 9 shows an embodiment of the present invention wherein the INA 149 functions as the differential amplifier of the circuit. Figure 10 shows another embodiment of the present invention wherein LTC5535 is used as the RF detecting component that could operate between 300 MFIz-7 GHz microwave frequency range at 5V operating voltage. The preferred embodiment of this invention is described in the above-mentioned detailed description. It is understood that those skilled in the art may conceive modifications and/or variations to the embodiment shown and described therein. Any such modifications or variations that fall within the purview of this description are intended to be included therein as well. Unless specifically noted, it is the intention of the inventors that the words and phrases in the specification and claims be given the ordinary and accustomed meanings to those of ordinary skill in the applicable art. The foregoing description of a preferred embodiment and best mode of the invention known to the applicant at the time of filing the application has been presented and is intended for the purposes of illustration and description. It is not intended to be exhaustive or to limit the present invention to the precise form disclosed, and many modifications and variations are possible in the light of the above teachings.

Claims

1. A handheld device for determining the variety and maturity level of cacao pods, the apparatus comprising: a chassis; a microwave frequency circuit built inside the chassis and aimed away from user; a battery management system; a handle connecting the microwave frequency circuit and the battery management system; a multifunction trigger disposed at the handle; an image capturing module disposed away from the user; a microprocessor connected with the microwave frequency circuit and the image capturing module; and a screen for displaying the variety and maturity level assessment from the microprocessor.

2. The handheld device according to claim 1 wherein the microwave frequency circuit comprises a microwave frequency oscillator; a resonator communicating with the microwave frequency oscillator for emitting the microwave frequency signal; an antenna covering the resonator and directing the microwave frequency signal towards the cacao pod; a detector for receiving a signal reflected from the cacao pod; and a differential amplifier connected with the detector for comparing voltage inputs to be sent to the microprocessor.

3. The handheld device according to claim 1 wherein the microprocessor further comprises a convolutional neural network for aiding in the processing of the cacao pod images.

4. The handheld device according to claim 1 wherein the multifunction trigger is a single-pole, single-throw (SPST) push button switch.

5. The handheld device according to claim 1 wherein the screen is a thin- film-transistor liquid crystal display (TFT LCD).

6. The handheld device according to claim 1 wherein the chassis is made up of 3D-printing filament.

7. The handheld device according to claim 3 wherein the convolutional neural network includes a classification of cacao varieties comprising BR-25, UF- 18, and PBC 123.

8. A method of determining the variety and maturity level of cacao pods, the method comprising:

— producing a microwave frequency through an oscillator;

— directing signal from said microwave frequency towards the pod through a resonator;

— penetrating through the pod with the said signal;

— receiving a reflected signal from the pod through the resonator;

— detecting voltage of the reflected signal through an RF detector;

— comparing the voltage from the RF detector with a reference DC voltage through a differential amplifier;

— capturing image of the cacao pod;

— analyzing and evaluating output from the differential amplifier and the image capturing module in the microprocessor; and

— displaying results through a display screen.