First Layer Optimization Of Convolutional Neural Networks For Iot Edge Devices

• doi: 10.1109/ICM63406.2024.10815744
• title: First Layer Optimization of Convolutional Neural Networks for IoT Edge Devices
• publisher: IEEE
• isbn: 979-8-3503-7940-2
• issn: 2332-7014
• rank: 2497
• access_type: LOCKED
• content_type: Conferences
• abstract: Optimizing Convolutional Neural Networks (CNNs) for hardware deployment has gained significant attention as the number of edge devices continues to grow. This is driven by the realization that edge computation reduces bandwidth usage and, consequently, overall power consumption. The first CNN layer, in particular, is receiving increased focus due to its direct impact on the network's overall accuracy. This paper provides an overview and summary of various first-layer optimization techniques. Additionally, a Hardware-Aware Technique (HAT) is proposed and evaluated for its potential as a first-layer optimization method. A comparison between the proposed HAT and other leading techniques, based on overall network accuracy and the quality of first-layer feature maps, demonstrates that the HAT is a strong candidate for first-layer CNN optimization. In a custom CNN architecture using a dataset acquired by a CMOS image sensor (CIS), the proposed HAT achieves a validation accuracy of 96.49%, which is highly competitive with other state-of-the-art approaches.
• article_number: 10815744
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10815744
• html_url: https://ieeexplore.ieee.org/document/10815744/
• abstract_url: https://ieeexplore.ieee.org/document/10815744/
• publication_title: 2024 International Conference on Microelectronics (ICM)
• conference_location: Doha, Qatar
• conference_dates: 14-17 Dec. 2024
• publication_number: 10815720
• is_number: 10815705
• publication_year: 2024
• publication_date: 14-17 Dec. 2024
• start_page: 1
• end_page: 5
• citing_paper_count: 0
• citing_patent_count: 0
• download_count: 66
• insert_date: 20241231
• index_terms:
  • ieee_terms:
    • Accuracy
    • Power demand
    • Image edge detection
    • Optimization methods
    • Computer architecture
    • Bandwidth
    • CMOS image sensors
    • Hardware
    • Microelectronics
    • Convolutional neural networks
  • author_terms:
    • Convolutional Neural Networks
    • Optimization techniques
    • Hardware Aware-Technique
    • Feature maps
  • dynamic_index_terms:
    • Neural Network
    • Convolutional Neural Network
    • Internet Of Things
    • Edge Devices
    • Internet Of Things Edge
    • Convolutional Neural Network Optimization
    • Power Consumption
    • Feature Maps
    • Validation Accuracy
    • Convolutional Neural Network Layers
    • Accuracy Of Network
    • Overview Of Techniques
    • Root Mean Square Error
    • Root-mean-square Deviation
    • RMSD
    • RMSE
    • Mean Square Error
    • Mean Square Deviation
    • Activation Function
    • Convolutional Layers
    • Backpropagation
    • Convolution Operation
    • Weights Of Layer
    • System Power Consumption
    • Convolutional Neural Network Implementation
    • Subsequent Layers
    • Validation Phase
    • Neural Network Techniques
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 979-8-3503-7940-2, isbnType: New-2005
  • format: Electronic ISBN, value: 979-8-3503-7939-6, isbnType: New-2005
• authors:
  • Author Name: Sajna Tasneem Shoukath Ali
    Affiliation: College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar
    Author URL: https://ieeexplore.ieee.org/author/970533116647251
    ID: 970533116647251
    Order: 1
    Author Affiliations:
    • College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar
  • Author Name: Abubakar Abubakar
    Affiliation: College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar
    Author URL: https://ieeexplore.ieee.org/author/592858237964846
    ID: 592858237964846
    Order: 2
    Author Affiliations:
    • College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar
  • Author Name: Arshad Khan
    Affiliation: College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar
    Author URL: https://ieeexplore.ieee.org/author/37088586592
    ID: 37088586592
    Order: 3
    Author Affiliations:
    • College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar
  • Author Name: Amine Bermak
    Affiliation: College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar
    Author URL: https://ieeexplore.ieee.org/author/37275353200
    ID: 37275353200
    Order: 4
    Author Affiliations:
    • College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar

Image Sensor

• sensor_type: CMOS
• resolution: H × V (specific resolution not provided)
• dynamic_range: not specified
• pixel_size: not specified
• dark_current: not specified

Optical Data

• focal_length: not specified
• aperture: not specified
• field_of_view: not specified
• distortion: not specified

Performance Metrics

Applications & Benefits

• cell_imaging: CMOS image sensors are used in various applications, including medical devices for imaging cellular structures.
• benefits: Benefits include low power consumption, high integration capability, and suitability for various applications.

Supporting Organizations

• supported_by: Qatar Research Development and Innovation Council

Manuscript Details

• publication_date: 14-17 Dec. 2024

Relevancy Score

• score: 9
• missing_fields:
  • dynamic_range
  • pixel_size
  • dark_current
  • focal_length
  • aperture
  • field_of_view
  • distortion
  • frame_rate
  • signal_to_noise_ratio
  • sensitivity
  • shutter_speed
  • power_consumption
  • noise

Processed JSON Filename

request_adbc1761-b829-4e60-816e-c2c4da1ebfa1-first_layer_optimization_of_convolutional_neural_networks_for_iot_edge_devices.json