Numerous studies have used pedometers, accelerometers or global positioning system receivers to describe human walking pattern. However, description of walking pattern mainly relied on analyzing epochs ≥ 1 min, whereas it has been shown that most of walking and stopping bouts last ≤ 30 sec in free-living conditions. The aim of this study was to implement and determine the accuracy of a specific data processing methodology for different wearable monitors in order to identify short walking and stopping bouts during outdoor walking sessions. Twenty healthy subjects performed different prescribed outdoor walking protocols at a ''usual'' and then at a ''slow'' pace, including both walking and stopping bouts with durations varying from 3 to 50 seconds (n= 768 bouts). Prescribed outdoor walking protocols were performed in two locations, corresponding to two distinct environmental conditions with low (outdoor athletic track, n=10 subjects) and high (urban canyon, n=10 subjects) levels of obstruction. All subjects wore one GPS DG100 with an external antenna located at the scapula (f=1 Hz), three GPS Qstarz BT-Q1000XT (f=1 Hz) and three Qstarz BT-Q1000EX (f=10 Hz) located at three positions (hip, wrist and scapula), as well as two wGT3X+ accelerometers placed at the hip and the wrist, giving each two types of data, namely, the counts (f=1Hz) and the raw data in G-Force (f=30Hz). The detection of the walking/stopping bouts was performed using the watershed method, which finds a threshold that best divides the data between two main states that are the walking and stopping states. From the entire set of subjects, 10 subjects (5 from each environment) were randomly selected to develop the method and the remaining 10 subjects were considered to test it. Accuracy was expressed in percentage with 95% confidence intervals. Among the 1Hz GPS devices, the DG100 SCAPULA and QStarz WRIST showed the highest performances with an accuracy of 96.0% (95.3-97.9) and 95.7% (96.0-98.4), respectively, while the QStarz HIP presented the worst accuracy [87.7% (86.1-91.0)]. Using a 10 Hz QStarz GPS receiver had no value in detecting walking/stopping bouts. Accuracies were higher for GPS data in environmental conditions with low levels of obstruction. Using counts from the wGT3X+ HIP provided an accuracy of 98.4% (96.5-98.8) in detecting walking and stopping bouts. The use of accelerometer's raw data had no value. Using a specific processing methodology, accelerometer and GPS monitors have the potential to detect short walking and stopping bouts with very high accuracy.