Detection, quantification and classification of ripened tomatoes: a comparative analysis of image processing and machine learning

Abstract

In this paper, specifically for detection of ripe/unripe tomatoes with/without defects in the crop field, two distinct methods are described and compared. One is a machine learning approach, known as 'Cascaded Object Detector' and the other is a composition of traditional customized methods, individually known as 'Colour Transformation', 'Colour Segmentation' and 'Circular Hough Transformation'. The (Viola Jones) Cascaded Object Detector generates 'histogram of oriented gradient' (HOG) features to detect tomatoes. For ripeness checking, the RGB mean is calculated with a set of rules. However, for traditional methods, color thresholding is applied to detect tomatoes either from a natural or solid background and RGB colour is adjusted to identify ripened tomatoes. In this work, Colour Segmentation is applied in the detection of tomatoes with defects, which has not previously been applied under machine learning techniques. The function modules of this algorithm are fed formatted images, captured by a camera mounted on a mobile robot. This robot was designed, built and operated in a tomato field to identify and quantify both green and ripened tomatoes as well as to detect damaged/blemished ones. This algorithm is shown to be optimally feasible for any micro-controller based miniature electronic devices in terms of its run time complexity of O(n 3) for traditional method in best and average cases. Comparisons show that the accuracy of the machine learning method is 95%, better than that of the Colour Segmentation Method using MATLAB. This result is potentially significant for farmers in crop fields to identify the condition of tomatoes quickly.

Dampak Keahlian Robot Dalam Mengenali Kematangan Tomat Terhadap Kepuasan Pengguna Pada Sektor Agroindustri

“…Proses pengawasan kematangan buah tomat sangat penting dalam produksi dan distribusi buah tomat yang berkualitas [6], [5]. Jika tomat dapat secara otomatis diklasifikasikan tingkat kematangannya menggunakan alat, hal ini akan mempermudah petani dalam meminimalkan pekerjaan manual yang membutuhkan waktu banyak [7], [8]. Selain itu, penghitungan jumlah tomat secara otomatis di lapangan akan meningkatkan efisiensi dalam menentukan nilai ekonomi, sehingga mampu meningkatkan produktivitas dalam pertanian [9].…”

Multi-Disease Recognition in Tomato Plants: Evaluating the Performance of CNN and Improved YOLOv7 Models for Accurate Detection and Classification

“…Since green tomatoes and backgrounds are so visually similar, little research has been done on the issue of green tomato recognition. This is shown by the research of Siddiquee et al [9]. They tried out a system that uses the cascaded object detector, a machine learning technique, in conjunction with more traditional image processing methods like "color trans-formation," "color segmentation," and "circular Hough transformation" to identify ripe tomatoes.…”

Evaluating the Single-Shot MultiBox Detector and YOLO Deep Learning Models for the Detection of Tomatoes in a Greenhouse

“…The problem of green tomato detection is rarely studied due to the difficulty of segmentation and differentiating it from the background, as they have similar colours. This can be observed by the comparison made by E Alam Siddiquee et al [33]. They compared a machine learning method, known as the "cascaded object detector", with a system that combines more traditional methods of image processing: "colour transformation", "colour segmentation", and "circular Hough transformation", in the detection of ripe tomatoes.…”