Employing the concept of stacking ensemble learning to generate deep dream images using multiple CNN variants
Addiction and adverse effects resulting from schizophrenia are rapidly becoming a global issue, necessitating the development of advanced approaches that can provide support to psychiatrists and psychologists to understand and replicate the hallucinations and imagery experienced by patients. Such ap...
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Intelligent Systems with Applications |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2667305325000146 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832540373998632960 |
|---|---|
| author | Lafta Alkhazraji Ayad R. Abbas Abeer S. Jamil Zahraa Saddi Kadhim Wissam Alkhazraji Sabah Abdulazeez Jebur Bassam Noori Shaker Mohammed Abdallazez Mohammed Mohanad A. Mohammed Basim Mohammed Al-Araji Abdulkareem Z. Mohmmed Wasiq Khan Bilal Khan Abir Jaafar Hussain |
| author_facet | Lafta Alkhazraji Ayad R. Abbas Abeer S. Jamil Zahraa Saddi Kadhim Wissam Alkhazraji Sabah Abdulazeez Jebur Bassam Noori Shaker Mohammed Abdallazez Mohammed Mohanad A. Mohammed Basim Mohammed Al-Araji Abdulkareem Z. Mohmmed Wasiq Khan Bilal Khan Abir Jaafar Hussain |
| author_sort | Lafta Alkhazraji |
| collection | DOAJ |
| description | Addiction and adverse effects resulting from schizophrenia are rapidly becoming a global issue, necessitating the development of advanced approaches that can provide support to psychiatrists and psychologists to understand and replicate the hallucinations and imagery experienced by patients. Such approaches can also be useful for promoting interest in human artwork, particularly surrealist images. Accordingly, in the present, a stacking ensemble Deep Dream model was developed that aids psychiatrists and psychologists in addressing the challenge of mimicking hallucinations. The dream-like images generated in the present study possess an aesthetic quality reminiscent of surrealist art. For model development, a series of five pre-trained Convolutional Neural Network (CNN) architectures—VGG-19, Inception v3, VGG-16, Inception-ResNet-V2, and Xception were stacked in an ensemble learning approach to create Deep Dream images whereby the upper hidden layers of the architectures were activated, and the models were trained via the Adam optimizer. Performance of the proposed model was evaluated across three octaves to amplify the maximum possible patterns and features of the base image. The resulting dream-like images contain shapes that reflect elements from the ImageNet dataset on which the above pre-trained models were trained. Each of the base images was manipulated to generate various dreamed images, each one with three octaves, which were finally combined to construct the final image with its loss. Final Deep Dream image showed a loss of 47.5821, while still retaining some features from the base image. |
| format | Article |
| id | doaj-art-2012636d856c41eb9fed4736d7f31c0d |
| institution | Kabale University |
| issn | 2667-3053 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Intelligent Systems with Applications |
| spelling | doaj-art-2012636d856c41eb9fed4736d7f31c0d2025-02-05T04:32:49ZengElsevierIntelligent Systems with Applications2667-30532025-03-0125200488Employing the concept of stacking ensemble learning to generate deep dream images using multiple CNN variantsLafta Alkhazraji0Ayad R. Abbas1Abeer S. Jamil2Zahraa Saddi Kadhim3Wissam Alkhazraji4Sabah Abdulazeez Jebur5Bassam Noori Shaker6Mohammed Abdallazez Mohammed7Mohanad A. Mohammed8Basim Mohammed Al-Araji9Abdulkareem Z. Mohmmed10Wasiq Khan11Bilal Khan12Abir Jaafar Hussain13Department of Computer Engineering Techniques, Imam Ja'afar Al-Sadiq University, IraqDepartment of Computer Science, University of Technology, Baghdad, IraqDepartment of Computer Technology Engineering, Al-Mansour University College, Baghdad, IraqDepartment of Computer Engineering Techniques, Imam Ja'afar Al-Sadiq University, IraqDepartment of Computer Engineering Techniques, Imam Ja'afar Al-Sadiq University, IraqDepartment of Computer Techniques Engineering, Imam Alkadhim University College, Baghdad, IraqComputer Science Department, College of Computer Science and Information Technology, University of Al-Qadisiyah, Al Diwaniyah, IraqUniversity of Karbala, College of Computer Science and Information Technology, Computer Science Department, IraqComputer Science Department, University of Technology, Baghdad, IraqImam Ja'afar Al-Sadiq University, IraqBabylon Education Directorate, Ministry of Education, Babylon, IraqSchool of Computer Science and Mathematics, Liverpool John Moores University, UKSchool of Computer and Engineering, California State University San Bernardino, USA; Corresponding author.Department of Electrical Engineering, University of Sharjah, Sharjah, UAEAddiction and adverse effects resulting from schizophrenia are rapidly becoming a global issue, necessitating the development of advanced approaches that can provide support to psychiatrists and psychologists to understand and replicate the hallucinations and imagery experienced by patients. Such approaches can also be useful for promoting interest in human artwork, particularly surrealist images. Accordingly, in the present, a stacking ensemble Deep Dream model was developed that aids psychiatrists and psychologists in addressing the challenge of mimicking hallucinations. The dream-like images generated in the present study possess an aesthetic quality reminiscent of surrealist art. For model development, a series of five pre-trained Convolutional Neural Network (CNN) architectures—VGG-19, Inception v3, VGG-16, Inception-ResNet-V2, and Xception were stacked in an ensemble learning approach to create Deep Dream images whereby the upper hidden layers of the architectures were activated, and the models were trained via the Adam optimizer. Performance of the proposed model was evaluated across three octaves to amplify the maximum possible patterns and features of the base image. The resulting dream-like images contain shapes that reflect elements from the ImageNet dataset on which the above pre-trained models were trained. Each of the base images was manipulated to generate various dreamed images, each one with three octaves, which were finally combined to construct the final image with its loss. Final Deep Dream image showed a loss of 47.5821, while still retaining some features from the base image.http://www.sciencedirect.com/science/article/pii/S2667305325000146Deep dreamStacking ensembleImaginary hallucinationsInceptionXceptionDeep learning |
| spellingShingle | Lafta Alkhazraji Ayad R. Abbas Abeer S. Jamil Zahraa Saddi Kadhim Wissam Alkhazraji Sabah Abdulazeez Jebur Bassam Noori Shaker Mohammed Abdallazez Mohammed Mohanad A. Mohammed Basim Mohammed Al-Araji Abdulkareem Z. Mohmmed Wasiq Khan Bilal Khan Abir Jaafar Hussain Employing the concept of stacking ensemble learning to generate deep dream images using multiple CNN variants Intelligent Systems with Applications Deep dream Stacking ensemble Imaginary hallucinations Inception Xception Deep learning |
| title | Employing the concept of stacking ensemble learning to generate deep dream images using multiple CNN variants |
| title_full | Employing the concept of stacking ensemble learning to generate deep dream images using multiple CNN variants |
| title_fullStr | Employing the concept of stacking ensemble learning to generate deep dream images using multiple CNN variants |
| title_full_unstemmed | Employing the concept of stacking ensemble learning to generate deep dream images using multiple CNN variants |
| title_short | Employing the concept of stacking ensemble learning to generate deep dream images using multiple CNN variants |
| title_sort | employing the concept of stacking ensemble learning to generate deep dream images using multiple cnn variants |
| topic | Deep dream Stacking ensemble Imaginary hallucinations Inception Xception Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S2667305325000146 |
| work_keys_str_mv | AT laftaalkhazraji employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT ayadrabbas employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT abeersjamil employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT zahraasaddikadhim employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT wissamalkhazraji employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT sabahabdulazeezjebur employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT bassamnoorishaker employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT mohammedabdallazezmohammed employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT mohanadamohammed employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT basimmohammedalaraji employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT abdulkareemzmohmmed employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT wasiqkhan employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT bilalkhan employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants AT abirjaafarhussain employingtheconceptofstackingensemblelearningtogeneratedeepdreamimagesusingmultiplecnnvariants |