Description:TECHNICAL FIELD
[0001] The embodiments of the present disclosure generally relate to the field of automated inspection and evaluation technologies that involve the use of machines or systems to classify and grade objects. More particularly, the present disclosure relates to a system, apparatus, and method for automated grading of objects.

BACKGROUND
[0002] The following description of the related art is intended to provide background information pertaining to the field of the disclosure. This section may include certain aspects of the art that may be related to various features of the present disclosure. However, it should be appreciated that this section is used only to enhance the understanding of the reader with respect to the present disclosure, and not as admissions of the prior art.
[0003] Sorting of fruit, for example, pomegranates on farms is a vital step in guaranteeing that the fruit reaches markets in good condition and meets the diverse requirements of consumers. Sorting pomegranates into different quality tiers allows farmers to tailor their marketing and packaging strategies to target specific customer segments. This process involves separating pomegranates into multiple categories based on size, ripeness, and quality. Farmers and workers carefully inspect each pomegranate, assessing its external appearance and internal attributes. Pomegranates that are free of blemishes, fully ripe, and of uniform size are typically classified as premium quality and are destined for high-end markets or specialty stores. Meanwhile, those with minor imperfections or variations may be grouped as standard-grade pomegranates, suitable for general retail distribution. This sorting process not only ensures that consumers receive the best-quality pomegranates but also maximizes the market value of the produce, as premium-grade pomegranates often command higher prices, while standard-grade fruits cater to a broader consumer base.
[0004] The traditional method of manual sorting of pomegranates has several drawbacks. One of the most significant drawback is that it is labor-intensive. Sorting pomegranates by hand requires a large workforce, and the repetitive task can be physically demanding, leading to worker fatigue. This labor-intensive approach also contributes to increased production costs, as farms need to invest in a larger workforce, which affects the overall profitability of the operation. Human judgment can vary, leading to inconsistencies in sorting criteria such as size, ripeness, and quality. This variability can result in a mix of pomegranates within the same category, making it difficult to meet the quality standards demanded by consumers and markets.
[0005] Various studies describe machine learning and image processing techniques for fruit sorting, particularly pomegranates. The studies describe methods using hybrid models, deep learning, feature extraction techniques like Local Binary Pattern (LBP), Histogram of Oriented Gradients (HOG), and neural networks (Artificial Neural Network (ANN), Support Vector Machine (SVM), Convolutional Neural Network (CNN)) for classifying fruit quality based on size, ripeness, and external features. Several studies highlight the effectiveness of combining different algorithms, such as CNNs with SVMs to improve accuracy. Some focus on non-destructive methods for quality evaluation, while others emphasize the importance of feature sets, image pre-processing, and optimization strategies for real-time applications. These approaches aim to reduce labor costs, increase efficiency, and ensure consistent fruit quality in agricultural processes.
[0006] Patent document CN117380558A discloses a spherical fruit online grading device and method based on machine vision, where the spherical fruit online grading device based on machine vision includes a conveying device, a guide rod, a photoelectric sensor, an image acquisition device, a human-machine interface (HMI) unit, an upper computer, and a grading device. The upper computer is internally preset with a spherical fruit grading detection algorithm, and the industrial personal computer runs the deployed spherical fruit grading detection algorithm. Pose information is obtained by processing images and matching templates, and the industrial personal computer controls a robot to drive a pneumatic clamp to clamp fruits and put the fruits into the corresponding fruit collecting boxes, so that intelligent online grading of the fruits is realized.
[0007] Patent document US9789518B2 discloses a fruit sorting apparatus which includes a conveyance device configured to convey a plurality of charged fruits from a detection area located on an upstream side to a sorting area disposed on a downstream side. An imaging device is configured to photograph the fruits in the detection area of the conveyance device. A processor is configured to detect a rotten fruit and a nonstandard fruit having an irregular shape or size based on images of the fruits and vegetables imaged by the imaging device. Further, a robot is configured to pick up the rotten fruit and the nonstandard fruit detected by the processor in the sorting area.
[0008] Patent document CN113102277A discloses a sorting method and a sorting system for fresh products. The sorting method includes acquiring categories of products to be classified and uploading the categories to a server. The sorting method includes acquiring the weight and the image of a product to be classified, and uploading the weight and the image of the product to be classified to the server. The server determines the grade of the product to be classified by using a classification model corresponding to the grade of the product to be classified according to the weight and the image of the product to be classified. The sorting method includes receiving the grade of the product to be classified from the server, and sorting the product to be classified to the corresponding sorting area according to the grade and the grade of the product to be classified.
[0009] Patent document CN118314375A discloses a fruit quality grading method based on deep learning, where fruit images are acquired through the Internet and an image acquisition device, data enhancement is carried out, and targeted amplification is carried out on images of different categories, so that a network model is fully trained. Preprocessing of an image is performed by a Gaussian filtering method to avoid the influence of noise on model training. The contrast between the fruit defect area and the normal area is improved by gamma correction by adopting an image enhancement method. The image data set is divided into a training set, a verification set, and a test set according to the proportion, and separately stored according to the image types. A convolutional neural network model is constructed, structurally adjusting and optimizing the convolutional neural network model, and to obtain a network model with the best classification effect. The trained model is deployed to achieve sorting of fruits.
[0010] Therefore, there is a need for a system and a method that can mitigate the problems associated with conventional systems and provide an efficient solution for sorting objects.

OBJECTS OF THE PRESENT DISCLOSURE
[0011] Some of the objects of the present disclosure, which at least one embodiment herein satisfies are listed herein below.
[0012] It is an object of the present disclosure to provide a system, apparatus, and a method for automated grading of objects that receives images associated with an object and generates structured images associated with the object based on the received images.
[0013] It is an object of the present disclosure to extract features from the structured images associated with the object.
[0014] It is an object of the present disclosure to classify, via a machine learning engine, the structured images of the object into various grades based on the extracted features.
[0015] It is an object of the present disclosure to analyze one or more parameters like, but not limited to, a texture, a shape, and a colour associated with the object from the structured images and classify the structured images of the object into one or more visual grades.
[0016] It is an object of the present disclosure to subsequently classify the structured images of the object into size grades based on a diameter of the object.

SUMMARY
[0017] This section is provided to introduce certain objects and aspects of the present disclosure in a simplified form that are further described below in the detailed description. This summary is not intended to identify the key features or the scope of the claimed subject matter.
[0018] In an aspect, the present disclosure relates to an automated grading system. The automated grading system includes a processor and a memory operatively coupled with the processor, where said memory stores instructions which, when executed by the processor, cause the processor to receive images of an object and generate structured images of the object based on the received images. The processor extracts one or more features from the structured images of the object. The processor classifies, via a machine learning engine, the structured images of the object into one or more grades based on the extracted one or more features.
[0019] In an embodiment, to generate the structured images of the object, the processor may be configured to remove one or more background elements from the images to isolate the object from the one or more background elements. The processor may be configured to subsequently redistribute one or more pixel intensities associated with the images to generate uniform images of the object. The processor may be configured to recalculate the one or more pixel intensities associated with the uniform images to generate the uniform images of the object with a single intensity value.
[0020] In an embodiment, the one or more features of the object may include any or a combination of, a texture associated with the object, a shape associated with the object, and a colour associated with the object.
[0021] In an embodiment, to classify the structured images into the one or more grades, the processor may be configured to analyze the texture, the shape, and the colour associated with the object from the structured images and classify the structured images of the object into one or more visual grades. The processor may be configured to subsequently classify the structured images of the object into one or more size grades based on a diameter of the object.
[0022] In an embodiment, the one or more visual grades may include any or a combination of, a Mostly Raw (MR) visual grade, a Damaged (DM) visual grade, and a Sunburned (SB) visual grade.
[0023] In an embodiment, the processor may be configured to use an artificial neural network (ANN) technique at the machine learning engine to classify the structured images of the object into the one or more grades.
[0024] In an aspect, the present disclosure relates to an apparatus for grading of objects. The apparatus includes a conveyor belt to secure one or more objects, where a DC motor is electrically coupled to the conveyor belt for powering the conveyor belt. One or more servo motors are electrically coupled to the conveyor belt to sort the one or more objects. An image acquisition unit is electrically coupled to the conveyor belt to generate images associated with the one or more objects. A processor is communicatively coupled with the conveyor belt, the DC motor, the one or more servo motors, and the image acquisition unit, where the processor is configured to receive images associated with the one or more objects through the image acquisition unit. The processor is configured to generate structured images associated with the one or more objects based on the received images. The processor is configured to extract one or more features from the structured images associated with the one or more objects. The processor is configured to classify, via a machine learning engine, the structured images of the object into one or more grades based on the extracted one or more features. The processor is configured to sort the one or more objects through the one or more servo motors based on the classified one or more grades.
[0025] In an aspect, the present disclosure relates to a method for automated grading. The method includes receiving, by a processor, associated with a system, images associated with an object and generating structured images associated with the object based on the received images. The method includes extracting, by the processor, one or more features from the structured images associated with the object. The method includes classifying, by the processor, via a machine learning engine, the structured images of the object into one or more grades based on the extracted one or more features.
[0026] In an embodiment, for generating the structured images of the object, the method may include removing, by the processor, one or more background elements from the images for isolating the object from the one or more background elements. The method may include subsequently redistributing, by the processor, one or more pixel intensities associated with the images for generating one or more uniform images of the object. The method may include recalculating, by the processor, the one or more pixel intensities associated with the uniform images for generating the uniform images of the object with a single intensity value.
[0027] In an embodiment, for classifying the structured images into the one or more grades, the method may include analyzing, by the processor, a texture, a shape, and a colour associated with the object from the structured images and classifying the structured images of the object into one or more visual grades. The method may include subsequently classifying, by the processor, the structured images of the object into one or more size grades based on a diameter of the object.

BRIEF DESCRIPTION OF DRAWINGS
[0028] The accompanying drawings, which are incorporated herein, and constitute a part of this disclosure, illustrate exemplary embodiments of the disclosed methods and systems which like reference numerals refer to the same parts throughout the different drawings. Components in the drawings are not necessarily to scale, emphasis instead being placed upon clearly illustrating the principles of the present disclosure. Some drawings may indicate the components using block diagrams and may not represent the internal circuitry of each component. It will be appreciated by those skilled in the art that disclosure of such drawings includes the disclosure of electrical components, electronic components, or circuitry commonly used to implement such components.
[0029] FIG. 1 illustrates an example system architecture (100) of the proposed system (102), in accordance with an embodiment of the present disclosure.
[0030] FIG. 2 illustrates an example block diagram (200) of a proposed system (102), in accordance with an embodiment of the present disclosure.
[0031] FIG. 3 illustrates a flow diagram of an example method (300) implemented by the proposed system (102), in accordance with an embodiment of the present disclosure.
[0032] The foregoing shall be more apparent from the following more detailed description of the disclosure.

DETAILED DESCRIPTION
[0033] In the following description, for the purposes of explanation, various specific details are set forth in order to provide a thorough understanding of embodiments of the present disclosure. It will be apparent, however, that embodiments of the present disclosure may be practiced without these specific details. Several features described hereafter can each be used independently of one another or with any combination of other features. An individual feature may not address all of the problems discussed above or might address only some of the problems discussed above. Some of the problems discussed above might not be fully addressed by any of the features described herein.
[0034] The present disclosure describes an automated system for grading objects, e.g., pomegranates. The system captures multiple images of each pomegranates, uses feature extraction and machine learning with a processor to classify the pomegranates into 12 grades, following market standards. Three grades, Mostly Raw (MR), Damaged (DM), and Sunburned (SB) are based on visual factors, while the other nine are determined by size. The system integrates both software and hardware components for automated classification based on quality grades ( MR, DM, SB). The system involves data collection, preprocessing, and feature extraction, where pomegranate images are preprocessed through background removal using a library, contrast enhancement with Contrast Limited Adaptive Histogram Equalization (CLAHE) and grayscale conversion. Local Binary Pattern (LBP) is applied to extract texture features from the images, which are then normalized and fed into a feedforward artificial neural network (ANN) for classification. The dataset is split into training and testing sets, with the model trained using an optimizer and a categorical loss function. The model’s performance is evaluated using accuracy scores, classification reports, and confusion matrices, and saved for real-time deployment. In the hardware setup, the processor acts as the central control unit, interfacing with DC motors through motor drivers to power a conveyor belt and servo motors to sort the pomegranates. The camera continuously captures images of pomegranates moving on the conveyor belt, which are processed by the processor for classification. Based on the predicted grade, the appropriate servo motor diverts the pomegranate to its correct category, and the result is displayed via for real-time sorting. This fully automated system enables efficient, quality-based sorting of pomegranates in agricultural processes.
[0035] Various embodiments of the present disclosure will be explained in detail with reference to FIGs. 1-3.
[0036] FIG. 1 illustrates an example system architecture (100) of the proposed system (102), in accordance with an embodiment of the present disclosure.
[0037] In an embodiment, the system (102) may receive one or more images associated with an object and generate one or more structured images associated with the object based on the received one or more images.
[0038] In an embodiment, to generate the one or more structured images of the object, the system (102) may remove one or more background elements from the one or more images to isolate the object from the one or more background elements. The system (102) may subsequently redistribute one or more pixel intensities associated with the one or more images to generate one or more uniform images of the object. The system (102) may recalculate the one or more pixel intensities associated with the one or more uniform images to generate the one or more uniform images of the object with a single intensity value.
[0039] In an embodiment, the system (102) may extract one or more features from the one or more structured images associated with the object. The one or more features of the object may include but not limited to a texture associated with the object, a shape associated with the object, and a colour associated with the object. A Local Binary Pattern (LBP) may be applied by the system (102) to extract the one or more features from the one or more images, which may be further normalized.
[0040] In an embodiment, the system (102) may classify, via a machine learning engine, the one or more structured images of the object into one or more grades based on the extracted one or more features. The system (102) may use an artificial neural network (ANN) technique with the machine learning engine to classify, the one or more structured images of the object into the one or more grades. To classify the one or more structured images into the one or more grades, the system (102) may analyze the texture, the shape, and the colour associated with the object from the one or more structured images and classify the one or more structured images of the object into one or more visual grades. The one or more visual grades may include but not limited to a Mostly Raw (MR) visual grade, a Damaged (DM) visual grade, and a Sunburned (SB) visual grade. Further, the system (102) may subsequently classify the one or more structured images of the object into one or more size grades based on a diameter of the object.
[0041] In an embodiment, the system (102) may be part of an apparatus for grading of objects. The apparatus may include a conveyor belt to secure one or more objects, where a DC motor may be electrically coupled to the conveyor belt for powering the conveyor belt. One or more servo motors may be electrically coupled to the conveyor belt to sort the one or more objects. An image acquisition unit may be electrically coupled to the conveyor belt to generate one or more images associated with the one or more objects. A processor (202) may be communicatively coupled with the conveyor belt, the DC motor, the one or more servo motors, and the image acquisition unit, where the processor (202) may be configured to receive one or more images associated with the one or more objects through the image acquisition unit. The processor (202) may be configured to generate one or more structured images associated with the one or more objects based on the received one or more images. The processor (202) may be configured to extract one or more features from the one or more structured images associated with the one or more objects. The processor (202) may be configured to classify, via a machine learning engine, the one or more structured images of the object into one or more grades based on the extracted one or more features. The processor (202) may be configured to sort the one or more objects through the one or more servo motors based on the classified one or more grades. The apparatus may include a pipe adaptively coupled to the conveyor belt that gradually increases in size, allowing smaller objects to fall through earlier sections and larger ones to roll further, ensuring efficient and accurate size-based grading.
[0042] In an embodiment, the system (102) may be part of the apparatus for grading pomegranates. The system (102) may capture at least three images of each pomegranate and use feature extraction, and machine learning engine to classify the pomegranates into 12 grades, following Indian market standards. Three grades, Mostly Raw (MR), Damaged (DM), and Sunburned (SB) based on visual factors derived based on the classification the one or more structured images of the pomegranates into one or more visual grades, while the other nine grades may be determined by size of the pomegranates. The system (102) may subsequently classify the one or more structured images of the pomegranates into one or more size grades based on a diameter of the pomegranates.
[0043] In an exemplary embodiment, a 16 GB Raspberry Pi 5 may be implemented as the processor (202), interfacing with the DC motor through one or more L298N motor drivers to power the conveyor belt and the servo motors to sort the objects.
[0044] As illustrated in FIG. 1, in an embodiment, the system (102) may receive a dataset (104), where the dataset (104) may include the one or more images. The system (102) may include an image processing unit (106) that includes the CLAHE (108) and a background removal module (110). The CLAHE (108) may redistribute one or more pixel intensities associated with the one or more images to generate one or more uniform images of the object. The background removal module (110) may remove one or more background elements from the one or more images to isolate the object from the one or more background elements. Further, the system (102) may include a feature extraction module (112) that may extract a texture associated with the object, a shape associated with the object, and a colour associated with the object from the one or more structured images. The feature extraction module (112) may include the LBP (114) for extracting the one or more features from the one or more structured images. The system (102) may include a machine learning module (116) with the ANN technique (118) to classify the one or more structured images of the object into the one or more visual grades and the one or more size grades.
[0045] As illustrated in FIG. 1, in an embodiment, the system (102) may include a performance evaluation module (120) where one or more parameters may include but not limited to accuracy (122), F1-score (124), precision (126), and recall (128) associated with the object may be inspected. The system (102) may record files (130) through scalar elements (132) and a label encoder (134). The recorded file may include classified one or more structured images of the object based on the one or more visual grades and the one or more size grades. Further, the system (102) may operate the servo motor (136) to sort the object (138). The sorted objects may be displayed by the system (102) through a local host display (140).
[0046] In an embodiment, the system (102) offers an efficient, automated solution for sorting pomegranates, reducing the need for manual intervention and improving the consistency of grading. The integration of hardware and software components provides a scalable framework for similar applications in agricultural automation. Further, the system (102) eliminates human error, ensuring consistent and accurate results. The system (102) significantly reduces labour costs and speeds up the sorting process, providing a cost-effective and portable solution for farmers. By adhering to Indian grading standards, the system (102) enhances market acceptance and profitability while automating a traditionally labour-intensive task. The system (102) may be used by farmers, cooperatives, and agribusinesses for post-harvest sorting, aiding in export, local sales, and juicing. The system (102) offers a scalable solution for agricultural machinery manufacturers and creates opportunities for rural employment through deployment and maintenance. Unlike existing machines, which rely solely on weight-based grading, the system (102) identifies characteristics such as sunburn, virus damage, and mostly raw conditions, eliminating the need for manual visual inspection and providing a specialized solution tailored to object grading.
[0047] FIG. 2 illustrates an example block diagram (200) of a proposed system (102), in accordance with an embodiment of the present disclosure.
[0048] Referring to FIG. 2, the system (102) may comprise one or more processor(s) (202) that may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and/or any devices that process data based on operational instructions. Among other capabilities, the one or more processor(s) (202) may be configured to fetch and execute computer-readable instructions stored in a memory (204) of the system (102). The memory (204) may be configured to store one or more computer-readable instructions or routines in a non-transitory computer readable storage medium, which may be fetched and executed to create or share data packets over a network service. The memory (204) may comprise any non-transitory storage device including, for example, volatile memory such as random-access memory (RAM), or non-volatile memory such as erasable programmable read only memory (EPROM), flash memory, and the like.
[0049] In an embodiment, the system (102) may include an interface(s) (206). The interface(s) (206) may comprise a variety of interfaces, for example, interfaces for data input and output (I/O) devices, storage devices, and the like. The interface(s) (206) may also provide a communication pathway for one or more components of the system (102). Examples of such components include, but are not limited to, processing engine(s) (208) and a database (210), where the processing engine(s) (208) may include, but not be limited to, a data ingestion engine (212) and a machine learning engine (214).
[0050] In an embodiment, the processing engine(s) (208) may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement one or more functionalities of the processing engine(s) (208). In examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the processing engine(s) (208) may be processor-executable instructions stored on a non-transitory machine-readable storage medium and the hardware for the processing engine(s) (208) may comprise a processing resource (for example, one or more processors), to execute such instructions. In the present examples, the machine-readable storage medium may store instructions that, when executed by the processing resource, implement the processing engine(s) (208). In such examples, the system (102) may comprise the machine-readable storage medium storing the instructions and the processing resource to execute the instructions, or the machine-readable storage medium may be separate but accessible to the system (102) and the processing resource. In other examples, the processing engine(s) (208) may be implemented by electronic circuitry.
[0051] In an embodiment, the processor (202) may receive one or more images associated with an object through the data ingestion engine (212). The processor (202) may record one or more images in the database (210). The processor (202) may generate one or more structured images associated with the object based on the received one or more images.
[0052] In an embodiment, to generate the one or more structured images of the object, the processor (202) may remove one or more background elements from the one or more images to isolate the object from the one or more background elements. The processor (202) may subsequently redistribute one or more pixel intensities associated with the one or more images to generate one or more uniform images of the object. The processor (202) may recalculate the one or more pixel intensities associated with the one or more uniform images to generate the one or more uniform images of the object with a single intensity value.
[0053] In an embodiment, the processor (202) may extract one or more features from the one or more structured images associated with the object. The one or more features of the object may include but not limited to a texture associated with the object, a shape associated with the object, and a colour associated with the object. A Local Binary Pattern (LBP) may be applied by the processor (202) to extract the one or more features from the one or more images, which may be further normalized.
[0054] In an embodiment, the processor (202) may classify, via the machine learning engine (214), the one or more structured images of the object into one or more grades based on the extracted one or more features. The system (102) may use an artificial neural network (ANN) technique with the machine learning engine to classify, the one or more structured images of the object into the one or more grades. To classify the one or more structured images into the one or more grades, the processor (202) may analyze the texture, the shape, and the colour associated with the object from the one or more structured images and classify the one or more structured images of the object into one or more visual grades. The one or more visual grades may include but not limited to a Mostly Raw (MR) visual grade, a Damaged (DM) visual grade, and a Sunburned (SB) visual grade. Further, processor (202) may subsequently classify the one or more structured images of the object into one or more size grades based on a diameter of the object.
[0055] Further, in an embodiment, the processor (202) may be part of an apparatus (not shown) for grading of objects. The apparatus may include a conveyor belt to secure one or more objects, where a DC motor may be electrically coupled to the conveyor belt for powering the conveyor belt. One or more servo motors may be electrically coupled to the conveyor belt to sort the one or more objects. An image acquisition unit may be electrically coupled to the conveyor belt to generate one or more images associated with the one or more objects. The processor (202) may be communicatively coupled with the conveyor belt, the DC motor, the one or more servo motors, and the image acquisition unit, where the processor (202) may be configured to receive one or more images associated with the one or more objects through the image acquisition unit. The processor (202) may be configured to generate one or more structured images associated with the one or more objects based on the received one or more images. The processor (202) may be configured to extract one or more features from the one or more structured images associated with the one or more objects. The processor (202) may be configured to classify, via the machine learning engine (214), the one or more structured images of the object into one or more grades based on the extracted one or more features. The processor (202) may be configured to sort pomegranates through the one or more servo motors based on the classified one or more grades. The apparatus may include a pipe belt that gradually increases in size, adaptively coupled to the conveyor, allowing smaller objects to fall through earlier sections and larger ones to roll further, ensuring efficient and accurate size-based grading.
[0056] In an embodiment, the processor (202) may be configured with the apparatus for grading pomegranates. The processor (202) may capture one or more images of each pomegranate and use feature extraction, and machine learning engine to classify the pomegranates into, for example, 12 grades, following market standards. Three grades, Mostly Raw (MR), Damaged (DM), and Sunburned (SB) based on visual factors may be derived based on the classification of the one or more structured images of the pomegranates into one or more visual grades, while the other nine grades may be determined by size of the pomegranates. The processor (202) may subsequently classify the one or more structured images of the pomegranates into one or more size grades based on a diameter of the pomegranates.
[0057] FIG. 3 illustrates a flow diagram (300) of an example method (300) implemented by the proposed system (108), in accordance with an embodiment of the present disclosure.
[0058] As illustrated in FIG. 3, at step 302, the method (300) may include receiving, by a system (102), one or more images associated with an object and generating one or more structured images associated with the object based on the received one or more images. At step 304, the method (300) may include extracting, by the system (102), one or more features from the one or more structured images associated with the object. At step 306, the method (300) may include classifying, by the system (102), via a machine learning engine, the one or more structured images of the object into one or more grades based on the extracted one or more features.
[0059] While considerable emphasis has been placed herein on the preferred embodiments, it will be appreciated that many embodiments can be made and that many changes can be made in the preferred embodiments without departing from the principles of the disclosure. These and other changes in the preferred embodiments of the disclosure will be apparent to those skilled in the art from the disclosure herein, whereby it is to be distinctly understood that the foregoing descriptive matter is to be implemented merely as illustrative of the disclosure and not as a limitation.
ADVANTAGES OF THE INVENTION
[0060] The present disclosure provides an efficient, automated solution for sorting pomegranates, reducing the need for manual intervention, and improving the consistency of grading.
[0061] The present disclosure enables integration of hardware and software components providing a scalable framework for similar applications in agricultural automation.
[0062] The present disclosure provides an automated, compact, and affordable pomegranate grading machine designed to address inefficiencies in manual sorting, while reducing labour costs, eliminating inaccuracies, and improving profitability for farmers and vendors.
[0063] The present disclosure eliminates human error, ensuring consistent and accurate results. This significantly reduces labour costs and speeds up the sorting process, providing a cost-effective and portable solution for farmers.
[0064] The present disclosure enhances market acceptance and profitability while automating a traditionally labour-intensive task.
[0065] The present disclosure offers a scalable solution for agricultural machinery manufacturers and creates opportunities for rural employment through deployment and maintenance.
[0066] The present disclosure combines advanced machine learning with robust hardware, promoting innovation and improved agricultural productivity, aligning with sustainable development goals for technological advancement and enhanced supply chain management.
, Claims:1. An automated grading system (102), comprising:
a processor (202); and
a memory (204) operatively coupled with the processor (202), wherein said memory (204) stores instructions which, when executed by the processor (202), cause the processor (202) to:
receive a plurality of images of an object and generate one or more structured images based on the received one or more images of the object;
extract one or more features from the plurality of structured images of the object; and
classify, via a machine learning engine, the plurality of structured images of the object into one or more grades based on the extracted one or more features.

2. The system (102) as claimed in claim 1, wherein to generate the plurality of structured images of the object, the processor (202) is configured to:
remove one or more background elements from the plurality of images to isolate the object from the one or more background elements;
subsequently redistribute one or more pixel intensities associated with the plurality of images to generate uniform images of the object; and
recalculate the one or more pixel intensities associated with the uniform images to generate the uniform images of the object with a single intensity value.

3. The system (102) as claimed in claim 1, wherein the one or more features of the object comprise any or a combination of: a texture associated with the object, a shape associated with the object, and a colour associated with the object.

4. The system (102) as claimed in claim 3, wherein to classify the plurality of structured images into the one or more grades, the processor (202) is configured to:
analyze the texture, the shape, and the colour associated with the object from the plurality of structured images and classify the plurality of structured images of the object into one or more visual grades; and
subsequently classify the plurality of structured images of the object into one or more size grades based on a diameter of the object.

5. The system (102) as claimed in claim 4, wherein the one or more visual grades comprise any or a combination of: a Mostly Raw (MR) visual grade, a Damaged (DM) visual grade, and a Sunburned (SB) visual grade.

6. The system (102) as claimed in claim 1, wherein the processor (202) is configured to use an artificial neural network (ANN) technique at the machine learning engine to classify the plurality of structured images of the object into the one or more grades.

7. An apparatus for grading of objects, the apparatus comprising:
a conveyor belt to secure one or more objects, wherein a DC motor is electrically coupled to the conveyor belt for powering the conveyor belt;
one or more servo motors electrically coupled to the conveyor belt to sort the one or more objects;
an image acquisition unit electrically coupled to the conveyor belt to generate a plurality of images of the one or more objects;
a processor (202) communicatively coupled with the conveyor belt, the DC motor, the one or more servo motors, and the image acquisition unit, wherein the processor (202) is configured to:
receive the plurality of images of the one or more objects from the image acquisition unit;
generate a plurality of structured images of the one or more objects based on the received plurality of images;
extract one or more features from the plurality of structured images of the one or more objects;
classify, via a machine learning engine, the plurality of structured images of the one or more objects into one or more grades based on the extracted one or more features; and
sort the one or more objects through the one or more servo motors based on the classified one or more grades.

8. A method (300) for automated grading, the method (300) comprising:
receiving (302), by a processor (202), associated with a system (102), a plurality of images of an object and generating a plurality of structured images of the object based on the received plurality of images;
extracting (304), by the processor (202), one or more features from the plurality of structured images of the object; and
classifying (306), by the processor (202), via a machine learning engine, the plurality of structured images of the object into one or more grades based on the extracted one or more features.

9. The method (300) as claimed in claim 8, wherein for generating the plurality of structured images of the object, the method (300) comprises:
removing, by the processor (202), one or more background elements from the plurality of images for isolating the object from the one or more background elements;
subsequently redistributing, by the processor (202), one or more pixel intensities associated with the plurality of images for generating uniform images of the object; and
recalculating, by the processor (202), the one or more pixel intensities associated with the uniform images for generating the uniform images of the object with a single intensity value.

10. The method (300) as claimed in claim 8, wherein for classifying the plurality of structured images into the one or more grades, the method (300) comprises:
analyzing, by the processor (202), a texture, a shape, and a colour associated with the object from the plurality of structured images and classifying the plurality of structured images of the object into one or more visual grades; and
subsequently classifying, by the processor (202), the plurality of structured images of the object into one or more size grades based on a diameter of the object.