Counting crowds is essential for applications such as urban planning, traffic monitoring, and video surveillance. However, accurately counting the number of individuals (people) is challenging within an image, especially for real-time applications. Recently developed Dense Crowd Counting systems incur high computational costs due to the utilize of computation intensive deep networks. Accordingly, this work proposes a novel lightweight model named Lightweight Network for Dense Crowd (LDCrowdNet), specifically designed for crowd counting and crowd density estimation. LDCrowdNet integrates components such as a ShuffleNetV2 backbone network, a proposed novel enhanced ECANet for efficient channel attention, an introduced feature pyramid network (FPN), and a context module (CM) that serves as a Density Map Generator (DMG). Extensive experiments have been conducted across multiple benchmark datasets, illustrate that LDCrowdNet surpasses current state-of-the-art methods in terms of accuracy and robustness. This superiority is evident across multiple crowd counting datasets: ShanghaiTech, UCF_CC_50, and UCF-QNRF.