The Segmented Colour Feature Extreme Learning Machine: Applications in Agricultural Robotics

Abstract

This study presents the Segmented Colour Feature Extreme Learning Machine (SCF-ELM). The SCF-ELM is inspired by the Extreme Learning Machine (ELM) which is known for its rapid training and inference times. The ELM is therefore an ideal candidate for an ensemble learning algorithm. The Colour Feature Extreme Learning Machine (CF-ELM) is used in this study due to its additional ability to extract colour image features. The SCF-ELM is an ensemble learner that utilizes feature mapping via k-means clustering, a decision matrix and majority voting. It has been evaluated on a range of challenging agricultural object classification scenarios including weed, livestock and machinery detection. SCF-ELM model performance results were excellent both in terms of detection, 90 to 99% accuracy, and also inference times, around 0.01(s) per image. The SCF-ELM was able to compete or improve upon established algorithms in its class, indicating its potential for remote computing applications in agriculture.

Mask YOLOv7-Based Drone Vision System for Automated Cattle Detection and Counting

“…To perform a multi-stage counting in a UAV embedded with imagery sensors for images with complex features, a hybrid of template matching and spectral characteristics approach was improved by Sadgrove et al (2021). To obtain accurate results from the above-ground animal survey, thermal imagery and UAV were integrated for wildlife detection, segmentation, classification, and tracking by González et al (2018), and this was made possible by using a pixel with a specific threshold and binary mask that matches a template in different instances.…”

“…To obtain accurate results from the above-ground animal survey, thermal imagery and UAV were integrated for wildlife detection, segmentation, classification, and tracking by González et al (2018), and this was made possible by using a pixel with a specific threshold and binary mask that matches a template in different instances. The prospect in using computer vision for livestock detection from UAV imagery was revealed in Sadgrove et al (2021), this is in addition to different machine learning models that have been employed solely for detection and counting of animals from UAV imagery. Supervised pixel-based image classification method (Chabot et al, 2018) and unsupervised pixel-based image classification method (Han et al, 2019) have been used for animal identification and counting, and animal population overestimation, respectively, with proper preprocessing such as labeling and augmentation of the images for supervised data training method using either manual or automatic labeling method such as LabelMe (Russell et al, 2008) for effectual results.…”

An Image Classification Method of Unbalanced Ship Coating Defects Based on DCCVAE-ACWGAN-GP

“…Conventional coating defect detection is the inspection and judging of coating quality and the recording of the type and level of defects the staff produces within a specific time after the painting is completed. This method relies on domain experts or field technicians to identify the underlying visual features of defects, such as color [9][10][11][12], shape [13][14][15][16], and texture [17][18][19][20] through professional knowledge and practical experience, which not only increases the work intensity and work pressure of personnel but also greatly increases time cost, making it difficult to ensure the accuracy of coating defect identification [21] and the efficiency of coating operations [22]. With proposals based in ship manufacturing theory and the continuous popularization of artificial intelligence in the ship painting industry, people have begun to gradually apply intelligent technology to the understanding of ship painting defects, but there are fewer reports on the application of image-based ship coating defects recognition.…”

Robust Prescribed Trajectory Tracking Control of a Robot Manipulator Using Adaptive Finite-Time Sliding Mode and Extreme Learning Machine Method

“…Although, the output weights in ELM are generally computed by the least square optimization approach, in several studies, these parameters are obtained based on establishing the global stability [18,19]. Regarding the advantages of ELM, this method has been applied in different applications successfully, being proved as a promising approach in robotics and control [20][21][22].…”