Hobbyists who scrutinized reports to the FAA of alleged close calls with drones found that pilots reported near misses in only a small fraction of the cases, according to a study obtained by USA TODAY.

The study found that of the 764 close-call incidents between drones and other aircraft, only 27 were actually described by pilots as a "near miss."

Pilots reported taking evasive action 10 times, according to the study by the Academy of Model Aeronautics, an advocacy group for 180,000 hobbyists. The group scrutinized the reports that FAA released from Nov. 13, 2014, through Aug. 20 out of concern that hobbyists were unfairly blamed for drone safety problems.

None of the sightings involved a mid-air collision, after one suspected incident was debunked as a bird strike. Several crashes involved military drones, rather than the civilian and local-government drones that FAA regulates.

"What it reflects is a much more complex equation than merely calling these things close calls or near misses," Richard Hanson, the model group's government-affairs director, told USA TODAY. "We're hoping the report will put it back into perspective."

Drone sightings sparked widespread concerns in August because the FAA reported that the number was on pace to quadruple this year, from 238 in 2014. The release was titled: "Pilot Reports of Close Calls With Drones Soar in 2015."

Airline pilots are increasingly reporting drones while approaching airports, raising the specter of mid-air collisions that could hurt travelers. For example, pilots reported a cluster sightings in August at 2,000 feet while approaching New Jersey's Newark airport.

Firefighters in California temporarily suspended flights in June, which allowed wildfires to spread and cause more damage, after seeing drones near their aircraft at 10,000 feet.

And a small drone crashed Jan. 26 on White House grounds, sparking federal security concerns.

But the Academy of Model Aeronautics found that many drone sightings reported to FAA were vague, mistaken or involved remotely piloted aircraft that were following the rules.

The study found 27 cases, 3.5%, where the pilot reported a "near miss" or "near collision" or "NMAC," for near mid-air collision. The study also found several reports where the pilot "isn't reporting a near mid-air" or "did not consider it as a NMAC."

"I think it should be better defined where the problem lies," Hanson said.

Hobbyists are supposed to fly lower than 400 feet high and commercial drone operators 500 feet. All are supposed to fly during daylight hours within sight of the remote pilot, at least 5 miles from an airport and away from crowds, unless they have special permission from FAA.

The reports to FAA sometimes contained vague descriptions. A Qantas pilot reported a "mini blimp" near Los Angeles airport on May 31. An American Airlines pilot near New York's LaGuardia airport reported a "model rocket" on July 5. Someone landing at Andrews Air Force Base near Washington, D.C., reported seeing something that resembled a "large vulture" on Nov. 30, 2014.

A half-dozen cases involved sightings that sounded too high for off-the-shelf drones, at heights ranging from 19,000 feet near Beaumont, Texas, to 51,000 feet near Washington. The report called that highest sighting July 24 -- above where airliners typically fly -- a UFO, although the planet of origin remained uncertain.

Some sightings involved permitted flights. For example, a "low flying" drone was spotted 5 miles south of the airport in Allentown, Pa., on July 24. A 911 caller reported a "not high" drone 13 miles from Chicago's Midway airport on April 28.

The Los Angeles Police Department asked the Inglewood Police Department to remove its drone flying over a crime scene Aug. 18 because it was 2 miles from the end of a runway at Los Angeles International Airport.

More than a dozen incidents -- including two crashes into the ground -- inolved military drones rather than the commercial or local-government drones that the FAA grants permits to fly.

An MQ-1C Gray Eagle military drone crashed near Wilsona Gardens, Calif., on March 25. Controllers at MacDill Air Force Base in Tampa, Fla. reported a drone crashed near a perimeter fence March 3.

Also, a Swedish Air Force C-130 climbed to avoid a military MQ-9 Reaper drone near Victorville, Calif., on May 26.

Industry advocates worry about a patchwork of rules being created because the slow pace of federal regulation encourages state and local governments. FAA expects to complete rules for commercial drones weighing up to 55 pounds in mid-2016.

Meanwhile, California Gov. Jerry Brown vetoed a bill Thursday that would have prohibited drones from flying less than 350 feet above private property without the owner's permission. The bill would have greatly restricted drone flights because hobbyists fly up to 400 feet and commercial operators 500 feet.

"Before we go down that path, let's look at this more carefully," Brown said in his veto message.

Hanson said laws and regulations are already available to sanction reckless operators, but that misrepresenting drone problems leads to poor decisions.

"We would call it sort of a knee-jerk reaction," he said. "We're starting to see legislation that is well-intended, but ill-conceived."

The FAA threatens US$25,000 civil fines for reckless flying, with dozens of open cases, and relies on law-enforcement agencies for criminal charges. On Friday, University of Kentucky campus police charged law student Peyton Wilson, 24, of Louisville, with a misdemeanor for flying a drone that crashed into the remodeled Commonwealth Stadium before a football game Sep. 5.

But enforcement is difficult because authorities on the ground have trouble tracking drone operators from reports in the sky. Local law enforcement wasn't called or mentioned in 142 reports to FAA, according to the study.

Hanson said a "carrot-and-stick" approach requires better enforcement as a deterrent. But the hobbyist group contends that with more than 500,000 drones sold already, the few hundred sightings reported to FAA are a "small fraction" of flights.

"We believe the vast majority of people choose to operate safely and responsibly, and they'll do a better job if education with the right safety information," Hanson said.

The hobbyist group made several recommendations to FAA, urging removal of cases that aren't clearly remote-controlled aircraft and better investigation of results.

"Once you have better analysis, you can better decide on what actions to take, whether that becomes more educational, legislative or regulatory," Hanson said.