A Novel Segmentation Recognition Algorithm of Agaricus bisporus Based on Morphology and Iterative Marker-Controlled Watershed Transform

Abstract

Accurate recognition of Agaricus bisporus is a prerequisite for precise automatic harvesting in a factory environment. Aimed at segmenting mushrooms adhering together from the complex background, this paper proposes a watershed-based segmentation recognition algorithm for A. bisporus. First, the foreground of A. bisporus is extracted via Otsu threshold segmentation and morphological operations. Then, a preliminary segmentation algorithm and a novel iterative marker generation method are proposed to prepare watershed markers. On this basis, a marker-controlled watershed algorithm is adopted to segment and recognize A. bisporus individuals. All the algorithms are implemented based on OpenCV (Open Source Computer Vision) libraries. Tests on images of A. bisporus collected at the cultivation bed show that the average correct recognition rate of the proposed algorithm is 95.7%, the average diameter measurement error is 1.15%, and the average coordinate deviation rate is 1.43%. The average processing time is 705.7 ms per single image, satisfying the real-time constraints based on 1 image/s. The proposed algorithm performed better than the current Circle Hough Transform (OpenCV’s implementation). It is convenient and easy to operate, providing a sound basis for subsequent research on mechanized harvesting equipment for A. bisporus.

Improved YOLOv5-based image detection of cotton impurities

“…It is composed of a backbone network module for feature extraction by multilayer convolutional operations, a training module for dataset pre-training, a detection module for verifying the output results, and an inference deployment module for integrating the embedding into the programme. 7 YOLOv5 can demonstrate a new adaptive anchor frame calculation method for different scales of images from different datasets, scaling the input image to the input size required by the network itself, and then sequentially performing normalization and other operations to prepare the model for training. 8 YOLOv5 input is augmented with Mosaic data from YOLOv4 to improve the training speed and accuracy of the model.…”

An Improved YOLOv5s-Based Agaricus bisporus Detection Algorithm

“…We conducted a comparative analysis to evaluate the recognition performance of the proposed YOLOv5s-CBAM algorithm against several other algorithms for A. bisporus recognition, including the Submersion Method [4], Circle Hough Transform (CHT) [5], Marker-Controlled Watershed Transform (MCWT) [5], and Improved Segmentation Recognition Algorithm (ISRA) [35]. The comparison results are shown in Table 12.…”

“…Ji et al [4] proposed a "submersion method", which incorporated the depth information to effectively segment the adherent mushroom clusters and used the circle Hough Transform to measure the diameter of A. bisporus, achieving a recognition success rate of 92.37% and a diameter measurement error of 4.94%. Chen et al [5] proposed an A. bisporus segmentation recognition algorithm combining morphology and iterative markercontrolled watershed transformation, which achieved a high recognition success rate of 95.7% and a diameter measurement error of only 1.43%. Although the above algorithms have solved the A. bisporus detection problem to a certain extent, they rely on manual feature extraction and scene information.…”

Improved Real-Time Models for Object Detection and Instance Segmentation for Agaricus bisporus Segmentation and Localization System Using RGB-D Panoramic Stitching Images

“…Refs. [11,12] proposed a preliminary segmentation algorithm and a novel iterative label generation method for initial watershed marking. Although these methods achieved a 95.7% accuracy in the effective recognition of A. bisporus, significant errors persisted in the radius fitting of the mushrooms.…”