Find motion intervals above threshold

You are processing data from a single security camera.

You are given:

• A threshold value T (a real number).
• A vector (array) of time-stamped motion readings. Each reading is a pair [time, motion_level], where:
• time is a strictly increasing integer timestamp (in seconds),
• motion_level is a real number.

Whenever motion_level > T, we consider the camera to be "active". Consecutive active readings form a continuous active interval from the time of the first active reading to the time of the last active reading in that group.

Assume:

• The input vector is sorted by time in strictly increasing order.
• If two active readings are adjacent in the array (with no inactive reading between them), they belong to the same active interval, even if there is a time gap between their timestamps.
• If a reading is inactive (motion_level <= T), it does not belong to any active interval.

Return a list of active time intervals [start_time, end_time] covering all maximal contiguous segments of readings where motion_level > T.

Example:

• Input:

• readings = [[1, 0.4], [5, 0.2], [11, 0.9], [15, 0.9], [17, 0.8], [25, 0.5], [27, 0.8], [36, 0.9]]

• T = 0.8

• Output:

• [[11, 17], [27, 36]]

Explanation:

• Readings at times 11, 15, 17 have motion_level > 0.8, so they form one interval [11, 17].
• Readings at times 27 and 36 have motion_level > 0.8, forming interval [27, 36].
• All other readings have motion_level <= 0.8 and are ignored.

Implement a function that takes the readings and threshold and returns the list of such intervals.