<p>Imaging Flow Cytometry (IFC) combines the high-throughput capabilities of flow cytometry with morphological insights. However, commercial systems are prohibitively expensive, while low-cost alternatives typically suffer from motion blur and low signal-to-noise ratios (SNR), impeding the real-time detection of minute targets. To address these challenges, we propose a Acquisition-Adaptive Minute Target Detection Framework that establishes a hardware-software synergistic paradigm using off-the-shelf components. We introduce the SM-YOLO model, explicitly designed to compensate for imaging defects in high-speed flow environments. This architecture integrates Space-to-Depth Convolution (SPD-Conv) to prevent feature dispersion during downsampling and employs Monte Carlo Attention (MCAttn) to probabilistically suppress fluid background noise. Furthermore, a Mixed Depth Connection Strategy is utilized to resolve dense occlusions. Experimental results on a self-built dataset demonstrate that our framework achieves an inference speed of 1129.9 FPS with 94.5% mAP50, providing an approximately 4.5-fold computational redundancy over the high-speed monochrome camera acquisition rate (250 FPS). Furthermore, the system demonstrates robust generalization capabilities across varying noise levels, validating the proof-of-concept (PoC) feasibility of low-cost, high-precision high-throughput detection through efficient algorithmic compensation. This provides a scalable PoC framework for laboratory and future biomedical applications.</p>