Lehrstuhl für Angewandte Softwaretechnik
Chair for Applied Software Engineering

Sajjad Taheri                                                                                                       https://wwwbruegge.in.tum.de/lehrstuhl_1/images/linkedin.jpg
PhD candidate in computer science; researcher in machine
learning applications, in particular fine-grained visual classification
for industrial small parts and fasteners.
Email: sajjad.taheri (at) tum.de
Tel
: +49 (89) 289 18200
 
Technische Universität München
Institut für Informatik I1
Boltzmannstraße 3
85748 Garching b. München, Germany

 Office Hours:

  By appointment. Please contact me via email.

 Research Interests:

 Publications:

 Teaching:

 Projects:

 Theses and Guided Researches:

 Please find underneath the offered, in progress and finished theses topics. You can also write me an email, including your transcript of records and CV, in case you can't find enough information on this page.

Offered

Bachelor's Thesis
Getting Depth Image of Small Parts Using Multi-Cameras
Advisor
Sajjad Taheri
Author
 
Date
 
Since the small parts and fasteners are scale-variant (which means that with scaling them we'll end up having another small part), classification of them needs a fixed camera to ensure that the distance between the camera lens and the object is always fixed. However, using two or more cameras, we are able to get the depth information of the objects, including their size. In this bachelor thesis, you will work on getting these depth images, using two or more cameras. The true challenge is to consider the small parts characteristics (their small size and their shiny surface) and find solutions to handle them.
Master's Thesis
Detection of Unknown Classes in Classification of Small Parts
Advisor
Sajjad Taheri
Author
 
Date
 
Dealing with unknown objects in classification tasks is always of a great importance. By default, the classifier predicts them as one of the already known classes with a certain confidence score. However, in sensitive applications, this could reduce the performance of the system. In this master thesis, you will explore different approaches to detect unknown objects in classification of small parts (for instance, adding an unknown class with the images of multiple unknown items) and develop them using convolutional neural networks.
Important: You need a solid knowledge of deep learning/convolutional neural networks and experience programming with python.
 
 
Bachelor's Thesis
Using Traditional Methods to Classify the Small Parts
Advisor
Sajjad Taheri
Author
 
Date
 
Although Convolutional Neural Networks has been used recently in image classification tasks and surpass other techniques, there are certain situations, where we can apply traditional methods and get optimal results; for example, when there is no variation in position and direction of the object in an image and the color/light conditions remain also the same. In this bachelor thesis, you will explore these traditional methods and build a classifier for industrial small parts using OpenCV.
Important: You need to have experience in python programming and OpenCV.
 
 

In Progress

Bachelor's Thesis
Classification of Diatoms Using Convolutional Neural Networks
Advisor
Sajjad Taheri
Author
Bettina Heigl
Date
15.04.2018
Creating dataset for the diatoms (small algae which can be found in all waters), train a model using convolutional neural nets to classify them, and compare the results with the traditional methods.
Master's Thesis
A Multi-view CNN Approach to Classify Nuts in Overhauling Processes
Advisor
Sajjad Taheri
Author
René Svartdal Birkeland
Date
15.05.2018
A new approach to create datasets for classification of different nuts, considering the inner threads and camera angle. The challenge is to preserve the size information of the small part, using a fixed camera and fixed distance to that, while pointing out the camera lens to the nut in specific angles that it can capture its threads, length and overall shape. Using this setup and also other pictures, you will train a model with convolutional neural network to classify different nuts.
Master's Thesis
Detection of Damaged Small Pieces in Overhauling Processes
Advisor
Sajjad Taheri
Author
Ralf Schönfeld
Date
15.04.2018
Creating a model to detect damaged small pieces (screws, bolts, nuts, pins, washers, etc.) in overhauling industrial machineries, using deep learning and convolutinal neural networks
Master's Thesis
Comparison between using 3D-Modeled and Manually-photographed Photos in Classification of Small Parts
Advisor
Sajjad Taheri
Author
Amr Abdelraouf
Date
15.03.2018
Studying methods to create datasets from 3D models and use them in training a model for small parts classifier. This method will be compared with the normal manually photographing approach in terms of usability of dataset creation and performance of the classifiers.

Finished

Master's Thesis
LeSRec: Using the Asymmetric Weight Allocation for a Learner Speech Recognition System
Advisor
Sajjad Taheri
Author
Gopala Krishna Char Cheidu Raghavendrachar
Date
15.11.2017
Implementing a Speech Recognition System, which can learn fro the user input. The idea is to build an application to be able to monitor the system performance an give it feedback regarding the recognized phrases. These sample, together with the corrected labels, will be used to retrain the model to improve the performance.
Bachelor's Thesis
Providing Training Dataset for Automatic Recognition of Small Pieces
Advisor
Sajjad Taheri
Author
Jonas Pfab
Date
15.03.2017
Study the current approaches to get the right dataset for deep learning processes and implementation of an application for automatic data augmentation to enrich the training dataset
Master's Thesis
Comparison between Cloud-based and Offline Speech Recognition Systems
Advisor
Sajjad Taheri
Author
Elma Gazetic
Date
15.04.2017
Study the popular offline open-source speech recognition systems and training/tuning them in order to compare with the cloud-based solutions

 

 

 

Cookies make it easier for us to provide you with our services. With the usage of our services you permit us to use cookies.
More information Ok