An Enhanced YOLOv5 Model for Greenhouse Cucumber Fruit Recognition Based on Color Space Features

Abstract

The identification of cucumber fruit is an essential procedure in automated harvesting in greenhouses. In order to enhance the identification ability of object detection models for cucumber fruit harvesting, an extended RGB image dataset (n = 801) with 3943 positive and negative labels was constructed. Firstly, twelve channels in four color spaces (RGB, YCbCr, HIS, La*b*) were compared through the ReliefF method to choose the channel with the highest weight. Secondly, the RGB image dataset was converted to the pseudo-color dataset of the chosen channel (Cr channel) to pre-train the YOLOv5s model before formal training using the RGB image dataset. Based on this method, the YOLOv5s model was enhanced by the Cr channel. The experimental results show that the cucumber fruit recognition precision of the enhanced YOLOv5s model was increased from 83.7% to 85.19%. Compared with the original YOLOv5s model, the average values of AP, F1, recall rate, and mAP were increased by 8.03%, 7%, 8.7%, and 8%, respectively. In order to verify the applicability of the pre-training method, ablation experiments were conducted on SSD, Faster R-CNN, and four YOLOv5 versions (s, l, m, x), resulting in the accuracy increasing by 1.51%, 3.09%, 1.49%, 0.63%, 3.15%, and 2.43%, respectively. The results of this study indicate that the Cr channel pre-training method is promising in enhancing cucumber fruit detection in a near-color background.

YOLO-plum: A high precision and real-time improved algorithm for plum recognition

“…The biggest advantage of YOLO is that it is extremely fast, which makes it a great advantage in real-time detection tasks. YOLOv5 algorithm [ 32 – 34 ] makes some improvements on the basis of YOLOv4 [ 35 ], tasks. YOLOv5 algorithm [ 32 – 34 ] makes some improvements on the basis of YOLOv4 [ 35 ], so that its speed and accuracy in target detection are greatly improved.…”

“…YOLOv5 algorithm [ 32 – 34 ] makes some improvements on the basis of YOLOv4 [ 35 ], tasks. YOLOv5 algorithm [ 32 – 34 ] makes some improvements on the basis of YOLOv4 [ 35 ], so that its speed and accuracy in target detection are greatly improved. In the model training stage, YOLOv5 is augmented with Mosaic data, and new ideas such as adaptive sight frame calculation and adaptive image scaling are proposed.…”

Color-Driven Object Recognition: A Novel Approach Combining Color Detection and Machine Learning Techniques

“…The assortment of information from the eleven research papers offers a careful handle of color detection and identification systems, explaining the numerous applications and approaches that they utilize. From traditional RGB variety models to state of the art strategies like Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), and K-Nearest Neighbours (KNN) algorithms, this examination feature the variety of color detection techniques [1][2][3][4][5][6][7][8][9][10][11]. The purposes are various and incorporate PC vision, picture handling, security, horticulture, and openness for those with visual disabilities.…”

“…Studies investigating the color detection of soil utilizing advanced image processing handling give significant bits of knowledge into the assessment of soil properties that are basic for agricultural tasks, highlighting the reasonable uses of color detection in horticulture [5]. Moreover, research on banana development discovery represents how variety acknowledgment might be utilized to decide natural product quality [6]. One normal subject all through a few articles is the utilization of color detection to further develop openness and help the people who are Curiously, one examination utilizes CNNs to distinguish hued objects for blind people, showing how computer vision might assist with regular undertakings and openness for those with visual debilitations [7].…”

“…These incorporate growing datasets, handling application-explicit troubles, and researching novel methodologies such coordinating unsupervised machine learning for color image segmentation. In light of everything, these examination distributions give a significant commitment to the area of color detection overall, showing its flexibility, helpful applications, and forthcoming social impact [1][2][3][4][5][6][7][8][9][10][11]. These…”