This master thesis project is carried out by one student at the Royal Institute of Technology in collaboration with Cybercom Group. The aim was to evaluate and compare system design strategies for fruit recognition in harvesting robots and the performance of supervised machine learning classification methods when applied to this specific task. The thesis covers the basics of these systems; to which parameters, constraints, requirements, and design decisions have been investigated. The framework is used as a foundation for the implementation of both sensing system, and processing and classification algorithms. A plastic tomato plant with fruit of varying maturity was used as a basis for training and testing, and a Kinect v2 for Windows including sensors for high resolution color-, depth, and IR data was used for image acquisition. The obtained data were processed and features of objects of interest extracted using MATLAB and a SDK for Kinect provided by Microsoft. Multiple views of the plant were acquired by having the plant rotate on a platform controlled by a stepper motor and an Ardunio Uno. The algorithms tested were binary classifiers, including Support Vector Machine, Decision Tree, and k-Nearest Neighbor. The models were trained and validated using a five fold cross validation in MATLABs Classification Learner application. Peformance metrics such as precision, recall, and the F1-score, used for accuracy comparison, were calculated. The statistical models k-NN and SVM achieved the best scores. The method considered most promising for fruit recognition purposes was the SVM. / Det här masterexamensarbetet har utförts av en student från Kungliga Tekniska Högskolan i samarbete med Cybercom Group. Målet var att utvärdera och jämföra designstrategier för igenkänning av frukt i en skörderobot och prestandan av klassificerande maskininlärningsalgoritmer när de appliceras på det specifika problemet. Arbetet omfattar grunderna av dessa system; till vilket parametrar, begränsningar, krav och designbeslut har undersökts. Ramverket användes sedan som grund för implementationen av sensorsystemet, processerings- och klassifikationsalgoritmerna. En tomatplanta i pplast med frukter av varierande mognasgrad användes som bas för träning och validering av systemet, och en Kinect för Windows v2 utrustad med sensorer för högupplöst färg, djup, och infraröd data anvöndes för att erhålla bilder. Datan processerades i MATLAB med hjälp av mjukvaruutvecklingskit för Kinect tillhandahållandet av Windows, i syfte att extrahera egenskaper ifrån objekt på bilderna. Multipla vyer erhölls genom att låta tomatplantan rotera på en plattform, driven av en stegmotor Arduino Uno. De binära klassifikationsalgoritmer som testades var Support Vector MAchine, Decision Tree och k-Nearest Neighbor. Modellerna tränades och valideras med hjälp av en five fold cross validation i MATLABs Classification Learner applikation. Prestationsindikatorer som precision, återkallelse och F1- poäng beräknades för de olika modellerna. Resultatet visade bland annat att statiska modeller som k-NN och SVM presterade bättre för det givna problemet, och att den sistnömnda är mest lovande för framtida applikationer.

3D LiDAR people detection and tracking applications rely on extracting individual people from the point cloud for reliable tracking. A recurring problem for these applications is under-segmentation caused by people standing close or interacting with each other, which in turn causes the system to lose tracking. To address this challenge, we propose Kernel Density Estimation Clustering with Grid (KDEG) based on Kernel Density Estimation Clustering. KDEG leverages a grid to save density estimates computed in parallel, finding cluster centers by selecting local density maxima in the grid. KDEG reaches a remarkable accuracy of 98.4%, compared to HDBSCAN and Scan Line Run (SLR) with 80.1% and 62.0% accuracy respectively. Furthermore, KDEG is measured to be highly efficient, with a running time similar to state-of-the-art methods SLR and Curved Voxel Clustering. To show the potential of KDEG, an experiment with a real tracking application on two people walking shoulder to shoulder was performed. This experiment saw a significant increase in the number of accurately tracked frames from 5% to 78% by utilizing KDEG, displaying great potential for real-world applications.  In parallel, we also explored HDBSCAN as an alternative to DBSCAN. We propose a number of modifications to HDBSCAN, including the projection of points to the groundplane, for improved clustering on human groups. HDBSCAN with the proposed modifications demonstrates a commendable accuracy of 80.1%, surpassing DBSCAN while maintaining a similar running time. Running time is however found to be lacking for both HDBSCAN and DBSCAN compared to more efficient methods like KDEG and SLR. / <p>Arbetet är gjort på plats i Tokyo på Chuo Universitet utan samverkan från Umeå Universitet såsom utbytesprogram eller liknande.</p><p>Arbetet är delvis finansierat av Scandinavia-Japan Sasakawa Foundation.</p><p>Arbetet gick inte under vanlig termin, utan började 2023/05/01 och slutade 2023/08</p>