针对目前带式输送机托辊轴承声信号增强方法存在过度降噪导致信号失真、适应性差，提取复杂声场特征效果较差等问题，提出一种基于直方图噪声估计和维纳滤波的带式输送机托辊轴承声信号增强方法。首先，计算轴承声信号各频带功率谱，构建功率谱直方图进行噪声谱估计，并对噪声谱进行一阶递归平滑；其次，利用噪声谱估计计算维纳滤波增益函数，得到维纳滤波降噪后的声信号；然后，对降噪后的声信号进行包络谱分析，通过对比实际测得的故障特征频率和理论的故障特征频率，实现对带式输送机托辊轴承故障诊断。实验和现场测试结果表明：基于直方图噪声估计和维纳滤波的带式输送机托辊轴承声信号增强方法得到的包络谱中含有明显的故障频率及其倍频成分，实验数据信噪比至少提升了1.14 dB，现场数据信噪比至少提升了1.04 dB，轴承故障特征提取效果较好，可实现环境噪声干扰严重下的带式输送机托辊轴承声信号增强。

To address the issues in existing acoustic signal enhancement methods for belt conveyor idler bearings, such as excessive noise reduction leading to signal distortion, poor adaptability, and ineffective extraction of complex sound field characteristics, a method based on histogram noise estimation and Wiener filtering was proposed. First, the power spectrum of each frequency band of acoustic signal was computed, and a power spectrum histogram was constructed to estimate the noise spectrum, followed by first-order recursive smoothing of the noise spectrum. Next, the estimated noise spectrum was used to calculate the Wiener filter gain function, obtaining the noise-reduced acoustic signal. Then, envelope spectrum analysis was performed on the denoised acoustic signal. By comparing the measured fault characteristic frequencies with theoretical fault characteristic frequencies, fault diagnosis of the belt conveyor idler bearing was achieved. Experimental and field test results show that the proposed acoustic signal enhancement method, based on histogram noise estimation and Wiener filtering, produces an envelope spectrum containing distinct fault characteristic frequencies and their harmonic components. The signal-to-noise ratio (SNR) of the experimental data improved by at least 1.14 dB, while the SNR of the field data improved by at least 1.04 dB. The method demonstrates good performance in extracting bearing fault features and can effectively enhance acoustic signals for belt conveyor idler bearings under severe environmental noise interference.