• B Ganga
• Lata B T
• Venugopal K R

Crowd analysis, especially the precise counting of individuals, is becoming crucial in fields like urban planning, transportation, and event management. The two main challenges are: accurately counting individuals in varying crowd sizes and monitoring crowd density. To address the above challenges, we have proposed the Crowd Counting Stability Analyzer (CCSA) architecture to enable crowd counting and stability analysis in a density map. The CCSA architecture is designed in two phases (i) Crowd Counting and (ii) Stability Analyzer. The crowd counting uses a Unet Generative Adver sarial Network (UGAN), composed of Uskip Connection, Unet Generator, and Discriminator. The crowd stability analyzer uses a Convolutional Neural Network (CNN) to classify the crowd density as very low, low, moderate, high, and very high in a density map. Additionally, to present crowd count details and their corresponding categories, we have constructed a new dataset known as ‘DataC’ having 48 images with 12 distinct categories. The observation is evaluated using Mean Square Error (MSE) and Mean Absolute Error (MAE) metrics on diverse scenes and densities in the ShanghaiTech (A, B), UCF CC 50, and DataC datasets. The experimental results show the model’s effectiveness over MSGAN with the MAE and MSE metrics on the above datasets.

Unet Generative Adversarial Network (UGAN), CNN, USkip Connection, Density Map, Crowd Stability

"CCSA: Crowd Counting with Stability Analysis using Adversarial Network and CNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 3, page no.l70-l83, March-2024, Available :http://www.jetir.org/papers/JETIR2403B09.pdf

"CCSA: Crowd Counting with Stability Analysis using Adversarial Network and CNN", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 3, page no. ppl70-l83, March-2024, Available at : http://www.jetir.org/papers/JETIR2403B09.pdf