The invention relates to a method, a device, electronic equipment and a storage medium for electrocardiogram signal detection and classification. The method comprises the following steps of: extracting a signal waveform from an electrocardiogram signal; acquiring a morphological characteristic and a deep characteristic of the signal waveform, wherein the morphological characteristic comprises anyone of a TR wave amplitude difference characteristic, a PR wave quantity ratio characteristic, an ST wave band characteristic and a P wave variation characteristic, and the deep characteristic comprises a depth characteristic and a hierarchy characteristic; inputting the morphological characteristic and the deep characteristic into a classifier; and acquiring a classification result output by theclassifier so as to obtain the signal type of the electrocardiogram signal, wherein the classification result is a result that the classifier performs classification according to the morphological characteristic and the deep characteristic, and the signal type comprises an atrial fibrillation rhythm, a non-atrial fibrillation abnormal rhythm, a normal sinus rhythm and noise. The method can accurately identify various types of abnormal rhythms, avoids the situation that abnormal rhythms of non-atrial fibrillation types such as tachycardia, bradycardia and arrhythmia are classified into atrial fibrillation types in error, and improves the accuracy of classification of electrocardiogram signals.