• Senthamizhselvi
• Saravanan

Visual place recognition (VPR) employing deep learning (DL) is a cutting-edge method that can significantly advance the domain of computer vision (CV). Leveraging deep neural networks (DNN), this technology allows machines to autonomously recognize and find particular landmarks or places within visual scenes with excellent speed and accuracy. VPR allows machines to recognize their environments, producing better decisions about their interactions and activities with the surroundings. This study designs an Intelligent Visual Place Recognition using Seagull Optimization Algorithm with Deep Transfer Learning (IVPR-SOADTL) method. The IVPR-SOADTL suggest an incorporated approach to improve performance of visual place recognition. We leverage a MixNet system for feature extraction, optimizing its hyperparameters employing the Seagull Optimization Algorithm (SOA), and use Manhattan Distance as the similarity measurement metric. For further improving the model's performance, we implements the SOA, a nature-inspired optimization method, to fine-tune the MixNet hyperparameters. Manhattan Distance is chosen as the similarity measurement metric for its capability to take both horizontal and vertical spatial relationships among feature vectors. We calculate our IVPR-SOADTL system on benchmark datasets and compare it against existing methods. The results exhibit that our combined architecture substantially increases accuracy of visual place recognition, outperforming existing approaches. An extensive comparison study stated the improved performance of the IVPR-SOADTL model over other methodologies.

Visual places recognition; Transfer learning; Deep learning; MixNet model; Feature extraction; Seagull Optimization Algorithm

"Intelligent Visual Place Recognition using Seagull Optimization Algorithm with Deep Transfer Learning Model", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 10, page no.f56-f66, October-2023, Available :http://www.jetir.org/papers/JETIR2310409.pdf

"Intelligent Visual Place Recognition using Seagull Optimization Algorithm with Deep Transfer Learning Model", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 10, page no. ppf56-f66, October-2023, Available at : http://www.jetir.org/papers/JETIR2310409.pdf