Search

The global orange industry constantly faces new technical challenges to meet consumer demands for quality fruits. Instead of traditional subjective fruit quality assessment methods, the interest in the horticulture industry has increased in objective, quantitative, and non-destructive assessment methods. Oranges have a thick peel which makes their non-destructive quality assessment challenging. This paper evaluates the potential of short-wave NIR spectroscopy and direct sweetness classification approach for Pakistani cultivars of orange, i.e., Red-Blood, Mosambi, and Succari. The correlation between quality indices, i.e., Brix, titratable acidity (TA), Brix: TA and BrimA (Brix minus acids), sensory assessment of the fruit, and short-wave NIR spectra, is analysed. Mix cultivar oranges are classified as sweet, mixed, and acidic based on short-wave NIR spectra. Short-wave NIR spectral data were obtained using the industry standard F-750 fruit quality meter (310–1100 nm). Reference Brix and TA measurements were taken using standard destructive testing methods. Reference taste labels i.e., sweet, mix, and acidic, were acquired through sensory evaluation of samples. For indirect fruit classification, partial least squares regression models were developed for Brix, TA, Brix: TA, and BrimA estimation with a correlation coefficient of 0.57, 0.73, 0.66, and 0.55, respectively, on independent test data. The ensemble classifier achieved 81.03% accuracy for three classes (sweet, mixed, and acidic) classification on independent test data for direct fruit classification. A good correlation between NIR spectra and sensory assessment is observed as compared to quality indices. A direct classification approach is more suitable for a machine-learning-based orange sweetness classification using NIR spectroscopy than the estimation of quality indices. [ABSTRACT FROM AUTHOR]

Active noise control (ANC) systems employing adaptive filters suffer from stability issues in the presence of impulsive noise. New impulsive noise control algorithms based on filtered-x recursive least square (FxRLS) algorithm are presented. The FxRLS algorithm gives better convergence than the filtered-x least mean square (FxLMS) algorithm and its variants but lacks robustness in the presence of high impulsive noise. In order to improve the robustness of FxRLS algorithm for ANC of impulsive noise, two modifications are suggested. First proposed modification clips the reference and error signals while, the second modification incorporates energy of the error signal in the gain of FxRLS (MGFxRLS) algorithm. The results demonstrate improved stability and robustness of proposed modifications in the FxRLS algorithm. However, another limitation associated with the FxRLS algorithm is its computationally complex nature. In order to reduce the computational load, a hybrid algorithm based on proposed MGFxRLS and normalized step size FxLMS (NSS-FXLMS) is also developed in this paper. The proposed hybrid algorithm combines the stability of NSS-FxLMS algorithm with the fast convergence speed of the proposed MGFxRLS algorithm. The results of the proposed hybrid algorithm prove that its convergence speed is faster than that of NSS-FxLMS algorithm with computational complexity lesser than that of FxRLS algorithm. [ABSTRACT FROM AUTHOR]

• A new approach for non-destructive hand held fruit maturity meter is investigated. • Direct maturity state classification (mature/immature) using a classifier. • Developed hardware is evaluated by on tree test samples of mango fruit of different season. • Maturity classification performs better than maturity index value estimation. Estimation of on-tree mango maturity is essential for the prediction of harvest time. Dry matter (DM) is a useful index in deciding mango maturity, and post-harvest quality. Existing NIR based maturity meters employ machine learning regressors to predict a maturity index value (such as DM, oBrix, or etc.) and then impose a hard threshold on predicted value to estimate maturity state of the fruit. In this paper, a new approach for non-destructive hand-held fruit maturity meter is investigated. The developed approach directly classifies the maturity state (mature/immature) using a classifier trained on maturity labels assigned through standard DM thresholds for investigated mango varieties. To develop the hardware of the device, a commercial-off-the-shelf development kit of NIR micro-spectrometer in the spectral range of 400–1100 nm was used with an embedded computational hardware, a micro-halogen lamp, a lithium battery, and a display. The application software (developed in C++) is designed to collect interactance spectra, remove noise, reduce dimensionality, and classify maturity state. Performance of the developed approach is evaluated by on-tree test samples of mango fruit of different season. Comparison of both the literature reported indirect maturity estimation and proposed direct maturity classification is conducted. The test results show that the maximum accuracy achieved using indirect maturity estimation using hard thresholds is 55.9%. Whereas direct maturity classification using KNN achieved 88.2% accuracy in predicting the maturity state (mature/immature) of the test mangoes. Overall results show that the developed DM mango maturity method has considerable potential to detect maturity state of mangoes in practical situations. [ABSTRACT FROM AUTHOR]

[Display omitted] • Vis/NIRS quantified sweetness of 'honey' melons. • Direct classification quantifies sweetness better than indirect classification. • With indirect classification the accuracy is 80.2%. • With direct classification the accuracy is upto 88.12% for two classes. Melons are nutritious, healthy, and one of the most eatable summer fruits in South Asia, especially in Pakistan. A melon is delicious if it is sweet, however, the gauge of its sweetness depends on the individual taste buds. In this paper, a direct sweetness classifier is proposed as a quantitative measure, to predict the sweetness of melon as opposed to indirect measure of soluble solid content (SSC/°Brix) based thresholding for near-infrared (NIR) spectroscopy. To provide guidance for fruit sweetness classification, sensory test was conducted, and sweetness standards were established as; very sweet (with °Brix over 10), sweet (with °Brix between 7 and 10), and flat (with °Brix below 7) class. NIR spectral data obtained using F-750 produce quality meter (310–1100 nm) was analyzed to build SSC prediction model and direct sweetness classification model. The best SSC model was obtained using multiple linear regression on second derivative of spectral data (for wavelength range 729–975 nm) with correlation coefficient = 0.93, and root mean square error = 1.63 on test samples. Sweetness of test samples were obtained using °Brix thresholding with an accuracy of 55.45% for three classes. The best direct sweetness classifier was obtained using K nearest neighbor (KNN) on second derivative of spectral data (for wavelength range 729–975 nm) with an accuracy of 70.3% for three classes on test samples. It was further observed that classification accuracy for sweet and flat melon can be improved by combining sweet and very sweet class samples into one 'satisfactory' class (with °Brix over 7). For °Brix thresholding-based classification the accuracy was improved to 80.2% and for KNN based direct sweetness classification the accuracy was improved to 88.12%. Extensive evaluation validates our argument that modeling a direct sweetness classifier is a better approach as compared to °Brix based thresholding for sweetness classification using NIR spectroscopy. [ABSTRACT FROM AUTHOR]

Maturity estimation is critical for harvesting fruit at the correct time which in turn results in good quality and longer shelf life of fruit. Researchers have explored several fruit maturity indicators over the past years including chronological, physical and biochemical (destructive) features. In recent years, various rapid non-destructive techniques along with different chemometric algorithms have been developed to estimate fruit maturity at the time of harvest. Among the non-destructive techniques, near infrared (NIR) spectroscopy has shown promising results for estimation of fruit harvest maturity. This paper reviews extensive work done on maturity estimation of various fruits such as apple, mango, grapes, peaches, pears, and melons using different NIR spectrometers in recent years. We observed that spectral information along with parameters measured through chemometric algorithms are used to develop NIR based fruit maturity indices estimator using machine learning techniques. The machine learning model for fruit maturity estimation for one variety may not directly applicable to other varieties of the same fruit and require tuning of the estimator or a new estimator with more sample readings of that variety. To validate, we used the pre-built predictive Kensington Pride (KP) mango model on Felix F-750 for local variety of mango. The results showed that the pre-build KP mango model is not valid for local varieties without re-calibration for these varieties. We observed that prediction models built using partial least squares (PLS) regression show promising results for apple, mango, grapes, melon and kiwifruit. Non-linear models such as least squares support vector machine (LSSVM), also give promising results for fruits such as pear and peach. We also observed that few dedicated portable instruments are developed and commercialized using NIR technology and have been tested in-field. The available devices are expensive for in-field fruit maturity estimation that constraint its use at large in developing countries. Democratization of NIR spectroscopy fruit maturity estimation requires the focus to be shifted from expensive handheld NIR device, which covers wide range of fruits, to a more economical photodiode-based fruit maturity estimation device that is calibrated for a particular fruit. [ABSTRACT FROM AUTHOR]

• Onboard plant counting framework for efficient early crop emergence estimation. • Direct estimation of plant counts from aerial images, eliminating the need for costly orthomosaic formation. • Evaluation of three methods based on time complexity and accuracy, with average F1 scores ranging from 0.947 to 0.967. Crop emergence estimation at early crop growth stages is becoming increasingly important for the long-term sustainability of natural resources. It helps farmers and agricultural stakeholders in the efficient allocation of resources like water, pesticides, and fertilizers. It can be used to estimate the yield and seed quality, identify the region of potential yield losses, and make future agriculture plans. These future agriculture plans can play a crucial role in ensuring maximum crop population and yield while utilizing the same limited land and natural resources. Most of the existing plant counting frameworks require offline processing of images with computationally expensive algorithms including the structure for motion and multiview stereo to develop an orthomosaic. This study proposed a tobacco plant counting framework that directly estimates counts from aerial images and has the potential for real-time applicability. It consists of three core modules: overlap detection, plant detection, and plant counting. The overlap detection module replaces the need for computationally expensive orthomosaic formation to avoid counting repetition by overlap masking based on only visual cues. Three different methods are evaluated as core modules for finding an optimal solution for plant counting based on time complexity and accuracy. In the first method after overlap detection, semantic segmentation with U-NET is employed as a plant detection module. For plant counting, we count the connected pixels classified as plants to estimate the crop count. In the second method after overlap detection, object detection using YOLOv7 is utilized as a plant detection module followed by simply counting each detected plant. In the third method, we utilize YOLOv7 for object detection, similar to the second method. However, we introduce the SORT (Simple Online and Realtime Tracking) algorithm for object tracking. This object tracking replaces the overlap detection module making it a real-time applicable method. For plant counting, we assess the number of tracked plants. The proposed algorithm is evaluated on two distinct tobacco fields. The high-resolution aerial data is collected from tobacco fields near Peshawar, Pakistan, and is human-labelled. The first and second methods show average F1 scores of 0.947 and 0.9667, respectively, whereas the third method has the potential for real-time applicability with an average F1 score of 0.967. [ABSTRACT FROM AUTHOR]

• A new method for rice variety identification. • The classifier takes both spectral and time series information of each pixel. • Spectral unmixing model is used for labelling each pixel. • Radiometric indices are used to increase classification accuracy. • Experimental results have exhibited an overall accuracy of 98.6%. In the agriculture sector food productivity, security, and sustainability, imposed challenges on farmers, regulatory bodies, and policymakers due to increasing demand and depleting natural resources and environmental concerns. Rice crop holds a prominent place in Pakistan's agriculture sector, it is not only consumed locally but also exported to many countries including China. Gathering crop information such as variety maps, yield estimation, or etc. can help farmers, regulatory bodies, policymakers, and rice mills in decision-making. In Pakistan, crop information is collected through manual field surveys that require a lot of human labor, are costly, and are time-consuming. One cannot ignore human error and bias in the process. A new framework for pixel classification is proposed that uses both spectral and time-series data of Sentinal-2 satellite for mapping two rice varieties "Basmati" and "IRRI, grown in Pakistan. The data were collected from twelve rice fields (approx. 307 acres) of different geographical locations at 16-time instances to cover the complete rice-growing season (May–October) in 2019. A linear spectral unmixing model is used to determine sub-pixel information of water, soil, and vegetation content, which is used for labeling each pixel for supervised learning. The input to our classifier is a 16 × 15 image formed using 15 spectral features (12 spectral bands and 3 radiometric indices) of 16 carefully selected different time instances for each pixel. The output is a pixel-level classification (semantic segmentation) of each pixel into Basmati, IRRI, and others (soil, water, etc.). Experimental results have exhibited an excellent overall accuracy of 98.6% with the proposed approach. The Basmati rice obtained higher accuracy of 99.7% as compared to IRRI rice with an accuracy of 95.2%. [ABSTRACT FROM AUTHOR]