Deep Learning-based Fast Grasp Planning for Robotic Bin-picking by Small Data Set without GPU

<div>This paper proposes a deep learning-based fast grasp detection method with a small dataset for robotic bin-picking. We consider the problem of grasping stacked up mechanical parts on a planar workspace using a parallel gripper. In this paper, we use a deep neural network to solve the problem with a single depth image. To reduce the computation time, we propose an edge-based algorithm to generate potential grasps. Then, a convolutional neural network (CNN) is applied to evaluate the robustness of all potential grasps for bin-picking. Finally, the proposed method ranks them and the object is grasped by using the grasp with the highest score. In bin-picking experiments, we evaluate the proposed method with a 7-DOF manipulator using textureless mechanical parts with complex shapes. The success ratio of grasping is 97%, and the average computation time of CNN inference is less than 0.23[s] on a laptop PC without a GPU array. In addition, we also confirm that the proposed method can be applied to unseen objects which are not included in the training dataset. </div>

Download Full-text

Deep Learning-based Fast Grasp Planning for Robotic Bin-picking by Small Data Set without GPU

10.36227/techrxiv.14384864 ◽

2021 ◽

Author(s):

SHOGO ARAI ◽

ZHUANG FENG ◽

Fuyuki Tokuda ◽

Adam Purnomo ◽

Kazuhiro Kosuge

Keyword(s):

Neural Network ◽

Deep Learning ◽

Computation Time ◽

Depth Image ◽

Training Dataset ◽

Small Data ◽

Data Set ◽

Grasp Planning ◽

Mechanical Parts ◽

Bin Picking

Download Full-text

Deep learning-based real-time detection of neurons in brain slices for in vitro physiology

Scientific Reports ◽

10.1038/s41598-021-85695-4 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Mighten C. Yip ◽

Mercedes M. Gonzalez ◽

Christopher R. Valenta ◽

Matthew J. M. Rowan ◽

Craig R. Forest

Keyword(s):

Neural Network ◽

Deep Learning ◽

Patch Clamp ◽

Convolutional Neural Network ◽

Brain Slice ◽

Interference Microscopy ◽

Small Data ◽

Patch Clamping ◽

Data Set ◽

Full Automation

AbstractA common electrophysiology technique used in neuroscience is patch clamp: a method in which a glass pipette electrode facilitates single cell electrical recordings from neurons. Typically, patch clamp is done manually in which an electrophysiologist views a brain slice under a microscope, visually selects a neuron to patch, and moves the pipette into close proximity to the cell to break through and seal its membrane. While recent advances in the field of patch clamping have enabled partial automation, the task of detecting a healthy neuronal soma in acute brain tissue slices is still a critical step that is commonly done manually, often presenting challenges for novices in electrophysiology. To overcome this obstacle and progress towards full automation of patch clamp, we combined the differential interference microscopy optical technique with an object detection-based convolutional neural network (CNN) to detect healthy neurons in acute slice. Utilizing the YOLOv3 convolutional neural network architecture, we achieved a 98% reduction in training times to 18 min, compared to previously published attempts. We also compared networks trained on unaltered and enhanced images, achieving up to 77% and 72% mean average precision, respectively. This novel, deep learning-based method accomplishes automated neuronal detection in brain slice at 18 frames per second with a small data set of 1138 annotated neurons, rapid training time, and high precision. Lastly, we verified the health of the identified neurons with a patch clamp experiment where the average access resistance was 29.25 M$$\Omega$$ Ω (n = 9). The addition of this technology during live-cell imaging for patch clamp experiments can not only improve manual patch clamping by reducing the neuroscience expertise required to select healthy cells, but also help achieve full automation of patch clamping by nominating cells without human assistance.

Download Full-text

Transfer Deep Learning for Dental and Maxillofacial Imaging Modality Classification: A Preliminary Study

Journal of Clinical Pediatric Dentistry ◽

10.17796/1053-4625-45.4.3 ◽

2021 ◽

Vol 45 (4) ◽

pp. 233-238

Author(s):

Lazar Kats ◽

Marilena Vered ◽

Johnny Kharouba ◽

Sigalit Blumer

Keyword(s):

Deep Learning ◽

Imaging Modality ◽

Daily Practice ◽

Automatic Classification ◽

Training Dataset ◽

Small Data ◽

Data Set ◽

X Ray ◽

Tel Aviv

Objective: To apply the technique of transfer deep learning on a small data set for automatic classification of X-ray modalities in dentistry. Study design: For solving the problem of classification, the convolution neural networks based on VGG16, NASNetLarge and Xception architectures were used, which received pre-training on ImageNet subset. In this research, we used an in-house dataset created within the School of Dental Medicine, Tel Aviv University. The training dataset contained anonymized 496 digital Panoramic and Cephalometric X-ray images for orthodontic examinations from CS 8100 Digital Panoramic System (Carestream Dental LLC, Atlanta, USA). The models were trained using NVIDIA GeForce GTX 1080 Ti GPU. The study was approved by the ethical committee of Tel Aviv University. Results: The test dataset contained 124 X-ray images from 2 different devices: CS 8100 Digital Panoramic System and Planmeca ProMax 2D (Planmeca, Helsinki, Finland). X-ray images in the test database were not pre-processed. The accuracy of all neural network architectures was 100%. Following a result of almost absolute accuracy, the other statistical metrics were not relevant. Conclusions: In this study, good results have been obtained for the automatic classification of different modalities of X-ray images used in dentistry. The most promising direction for the development of this kind of application is the transfer deep learning. Further studies on automatic classification of modalities, as well as sub-modalities, can maximally reduce occasional difficulties arising in this field in the daily practice of the dentist and, eventually, improve the quality of diagnosis and treatment.

Download Full-text

Effectiveness of transfer learning for enhancing tumor classification with a convolutional neural network on frozen sections

Scientific Reports ◽

10.1038/s41598-020-78129-0 ◽

2020 ◽

Vol 10 (1) ◽

Author(s):

Young-Gon Kim ◽

Sungchul Kim ◽

Cristina Eunbee Cho ◽

In Hye Song ◽

Hee Jin Lee ◽

...

Keyword(s):

Neural Network ◽

Deep Learning ◽

Convolutional Neural Network ◽

Transfer Learning ◽

Frozen Section ◽

Medical Center ◽

External Validation ◽

Model Performance ◽

Classification Model ◽

Training Dataset

AbstractFast and accurate confirmation of metastasis on the frozen tissue section of intraoperative sentinel lymph node biopsy is an essential tool for critical surgical decisions. However, accurate diagnosis by pathologists is difficult within the time limitations. Training a robust and accurate deep learning model is also difficult owing to the limited number of frozen datasets with high quality labels. To overcome these issues, we validated the effectiveness of transfer learning from CAMELYON16 to improve performance of the convolutional neural network (CNN)-based classification model on our frozen dataset (N = 297) from Asan Medical Center (AMC). Among the 297 whole slide images (WSIs), 157 and 40 WSIs were used to train deep learning models with different dataset ratios at 2, 4, 8, 20, 40, and 100%. The remaining, i.e., 100 WSIs, were used to validate model performance in terms of patch- and slide-level classification. An additional 228 WSIs from Seoul National University Bundang Hospital (SNUBH) were used as an external validation. Three initial weights, i.e., scratch-based (random initialization), ImageNet-based, and CAMELYON16-based models were used to validate their effectiveness in external validation. In the patch-level classification results on the AMC dataset, CAMELYON16-based models trained with a small dataset (up to 40%, i.e., 62 WSIs) showed a significantly higher area under the curve (AUC) of 0.929 than those of the scratch- and ImageNet-based models at 0.897 and 0.919, respectively, while CAMELYON16-based and ImageNet-based models trained with 100% of the training dataset showed comparable AUCs at 0.944 and 0.943, respectively. For the external validation, CAMELYON16-based models showed higher AUCs than those of the scratch- and ImageNet-based models. Model performance for slide feasibility of the transfer learning to enhance model performance was validated in the case of frozen section datasets with limited numbers.

Download Full-text

Orchard Mapping with Deep Learning Semantic Segmentation

Sensors ◽

10.3390/s21113813 ◽

2021 ◽

Vol 21 (11) ◽

pp. 3813

Author(s):

Athanasios Anagnostis ◽

Aristotelis C. Tagarakis ◽

Dimitrios Kateris ◽

Vasileios Moysiadis ◽

Claus Grøn Sørensen ◽

...

Keyword(s):

Neural Network ◽

Deep Learning ◽

Semantic Segmentation ◽

Automated Detection ◽

Aerial Images ◽

Training Dataset ◽

Field Boundary ◽

Different Seasons ◽

Detection And Localization ◽

Different Levels

This study aimed to propose an approach for orchard trees segmentation using aerial images based on a deep learning convolutional neural network variant, namely the U-net network. The purpose was the automated detection and localization of the canopy of orchard trees under various conditions (i.e., different seasons, different tree ages, different levels of weed coverage). The implemented dataset was composed of images from three different walnut orchards. The achieved variability of the dataset resulted in obtaining images that fell under seven different use cases. The best-trained model achieved 91%, 90%, and 87% accuracy for training, validation, and testing, respectively. The trained model was also tested on never-before-seen orthomosaic images or orchards based on two methods (oversampling and undersampling) in order to tackle issues with out-of-the-field boundary transparent pixels from the image. Even though the training dataset did not contain orthomosaic images, it achieved performance levels that reached up to 99%, demonstrating the robustness of the proposed approach.

Download Full-text

A Comparison Study of Mahalanobis-Taguchi System and Neural Network for Multivariate Pattern Recognition

Design Engineering, Parts A and B ◽

10.1115/imece2005-80029 ◽

2005 ◽

Cited By ~ 10

Author(s):

Jungeui Hong ◽

Elizabeth A. Cudney ◽

Genichi Taguchi ◽

Rajesh Jugulum ◽

Kioumars Paryani ◽

...

Keyword(s):

Neural Network ◽

Small Data ◽

Data Sets ◽

Comparison Study ◽

Data Set ◽

Set Size ◽

Breast Cancer Study ◽

Discriminant Ability ◽

Small Data Sets ◽

Multivariate Pattern

The Mahalanobis-Taguchi System is a diagnosis and predictive method for analyzing patterns in multivariate cases. The goal of this study is to compare the ability of the Mahalanobis-Taguchi System and a neural network to discriminate using small data sets. We examine the discriminant ability as a function of data set size using an application area where reliable data is publicly available. The study uses the Wisconsin Breast Cancer study with nine attributes and one class.

Download Full-text

Embedded Deep Learning Prototyping Approach for Cyber-Physical Systems: Smart LIDAR Case Study

Journal of Sensor and Actuator Networks ◽

10.3390/jsan10010018 ◽

2021 ◽

Vol 10 (1) ◽

pp. 18

Author(s):

Quentin Cabanes ◽

Benaoumeur Senouci ◽

Amar Ramdane-Cherif

Keyword(s):

Neural Network ◽

Deep Learning ◽

Computation Time ◽

Physical World ◽

Cyber Physical Systems ◽

Physical Systems ◽

Research Technology ◽

Standard Design ◽

Hardware Architectures ◽

Voxel Grid

Cyber-Physical Systems (CPSs) are a mature research technology topic that deals with Artificial Intelligence (AI) and Embedded Systems (ES). They interact with the physical world via sensors/actuators to solve problems in several applications (robotics, transportation, health, etc.). These CPSs deal with data analysis, which need powerful algorithms combined with robust hardware architectures. On one hand, Deep Learning (DL) is proposed as the main solution algorithm. On the other hand, the standard design and prototyping methodologies for ES are not adapted to modern DL-based CPS. In this paper, we investigate AI design for CPS around embedded DL. The main contribution of this work is threefold: (1) We define an embedded DL methodology based on a Multi-CPU/FPGA platform. (2) We propose a new hardware design architecture of a Neural Network Processor (NNP) for DL algorithms. The computation time of a feed forward sequence is estimated to 23 ns for each parameter. (3) We validate the proposed methodology and the DL-based NNP using a smart LIDAR application use-case. The input of our NNP is a voxel grid hardware computed from 3D point cloud. Finally, the results show that our NNP is able to process Dense Neural Network (DNN) architecture without bias.

Download Full-text

RISC-V Virtual Platform-Based Convolutional Neural Network Accelerator Implemented in SystemC

Electronics ◽

10.3390/electronics10131514 ◽

2021 ◽

Vol 10 (13) ◽

pp. 1514

Author(s):

Seung-Ho Lim ◽

WoonSik William Suh ◽

Jin-Young Kim ◽

Sang-Young Cho

Keyword(s):

Neural Network ◽

Deep Learning ◽

Network Model ◽

Neural Network Model ◽

Deep Neural Network ◽

System Level ◽

Neural Network Models ◽

Data Set ◽

Embedded Device ◽

Virtual Platform

The optimization for hardware processor and system for performing deep learning operations such as Convolutional Neural Networks (CNN) in resource limited embedded devices are recent active research area. In order to perform an optimized deep neural network model using the limited computational unit and memory of an embedded device, it is necessary to quickly apply various configurations of hardware modules to various deep neural network models and find the optimal combination. The Electronic System Level (ESL) Simulator based on SystemC is very useful for rapid hardware modeling and verification. In this paper, we designed and implemented a Deep Learning Accelerator (DLA) that performs Deep Neural Network (DNN) operation based on the RISC-V Virtual Platform implemented in SystemC in order to enable rapid and diverse analysis of deep learning operations in an embedded device based on the RISC-V processor, which is a recently emerging embedded processor. The developed RISC-V based DLA prototype can analyze the hardware requirements according to the CNN data set through the configuration of the CNN DLA architecture, and it is possible to run RISC-V compiled software on the platform, can perform a real neural network model like Darknet. We performed the Darknet CNN model on the developed DLA prototype, and confirmed that computational overhead and inference errors can be analyzed with the DLA prototype developed by analyzing the DLA architecture for various data sets.

Download Full-text

Automatic Handgun Detection with Deep Learning in Video Surveillance Images

Applied Sciences ◽

10.3390/app11136085 ◽

2021 ◽

Vol 11 (13) ◽

pp. 6085

Author(s):

Jesus Salido ◽

Vanesa Lomas ◽

Jesus Ruiz-Santaquiteria ◽

Oscar Deniz

Keyword(s):

Neural Network ◽

Deep Learning ◽

Convolutional Neural Network ◽

Video Surveillance ◽

Automatic Detection ◽

Public Spaces ◽

Detection Methods ◽

Training Dataset ◽

Average Precision ◽

Terrorist Acts

There is a great need to implement preventive mechanisms against shootings and terrorist acts in public spaces with a large influx of people. While surveillance cameras have become common, the need for monitoring 24/7 and real-time response requires automatic detection methods. This paper presents a study based on three convolutional neural network (CNN) models applied to the automatic detection of handguns in video surveillance images. It aims to investigate the reduction of false positives by including pose information associated with the way the handguns are held in the images belonging to the training dataset. The results highlighted the best average precision (96.36%) and recall (97.23%) obtained by RetinaNet fine-tuned with the unfrozen ResNet-50 backbone and the best precision (96.23%) and F1 score values (93.36%) obtained by YOLOv3 when it was trained on the dataset including pose information. This last architecture was the only one that showed a consistent improvement—around 2%—when pose information was expressly considered during training.

Download Full-text

Food Volume Estimation Based on Deep Learning View Synthesis from a Single Depth Map

Nutrients ◽

10.3390/nu10122005 ◽

2018 ◽

Vol 10 (12) ◽

pp. 2005 ◽

Cited By ~ 12

Author(s):

Frank Lo ◽

Yingnan Sun ◽

Jianing Qiu ◽

Benny Lo

Keyword(s):

Neural Network ◽

Deep Learning ◽

Point Cloud ◽

Volume Estimation ◽

Assessment System ◽

View Synthesis ◽

Depth Image ◽

3D Point Cloud ◽

Viewing Angle ◽

3D Point Clouds

An objective dietary assessment system can help users to understand their dietary behavior and enable targeted interventions to address underlying health problems. To accurately quantify dietary intake, measurement of the portion size or food volume is required. For volume estimation, previous research studies mostly focused on using model-based or stereo-based approaches which rely on manual intervention or require users to capture multiple frames from different viewing angles which can be tedious. In this paper, a view synthesis approach based on deep learning is proposed to reconstruct 3D point clouds of food items and estimate the volume from a single depth image. A distinct neural network is designed to use a depth image from one viewing angle to predict another depth image captured from the corresponding opposite viewing angle. The whole 3D point cloud map is then reconstructed by fusing the initial data points with the synthesized points of the object items through the proposed point cloud completion and Iterative Closest Point (ICP) algorithms. Furthermore, a database with depth images of food object items captured from different viewing angles is constructed with image rendering and used to validate the proposed neural network. The methodology is then evaluated by comparing the volume estimated by the synthesized 3D point cloud with the ground truth volume of the object items.

Download Full-text