Following a bench trial, defendant was convicted of armed robbery, MCL 750.529; MSA 28.797. He was sentenced to three to ten years in prison as an habitual offender, fourth offense, MCL 769.12; MSA 28.1084, and appeals as of right. We affirm.
Defendant’s conviction arose out of a robbery in a Little Caesar’s parking lot at approximately 6:00 P.M. on December 8, 1996, in the city of Detroit. As complainant Yvette Williams carried her pizza to her car, which was parked in the first space next to the restaurant, she felt a tug at her purse. As she opened the car door and was entering her car, she turned around and was face to face with a man holding a knife and demanding money. His face was approximately one foot from hers. The complainant gave him money *699from her purse, but had four to five minutes to look at his face because he then asked her for additional money, which she got from her pocket, asked about her earrings, asked and remarked about how scared she was, and then asked for a piece of her pizza. Meanwhile, off-duty police sergeant Jadié Settles saw the robbery in progress from his vehicle, drove into the parking lot, and confronted the suspect, who was twelve to twenty feet from him. The suspect threw his knife to the ground and ran from the scene.
On two subsequent occasions, one a few days after the incident and another approximately one month later, the complainant saw the suspect in the area, but she was unsuccessful in getting the police to respond. Then, three months later, the complainant again spotted the suspect in another area restaurant; she called the police, and defendant was arrested at the scene. Six months later, immediately before the start of trial, Settles told the prosecutor that defendant “looks like the same guy.” At trial, both the complainant and Settles identified defendant as the perpetrator of the robbery. Defendant admitted that he gave a bogus name and address to the police at the time of his arrest, but denied any involvement in the crime and claimed that the evidence was insufficient to convict him because he did not match the physical descriptions given to the police by the complainant and Settles on the night of the incident. The trial court found that the differences in the description of defendant given by the complainant and Settles were not significant enough to raise a doubt regarding their positive identification of defendant.
On appeal, as at trial, defendant raises only the issue of identification evidence, again arguing that it *700was insufficient to support his conviction of armed robbery. We disagree. When determining whether sufficient evidence has been presented to support a conviction, we view the evidence in the light most favorable to the prosecution to determine if a rational trier of fact could have found that the essential elements of a crime were proved beyond a reasonable doubt. People v Johnson, 460 Mich 720, 722-723; 597 NW2d 73 (1999). The credibility of identification testimony is a question for the trier of fact that we do not resolve anew. People v Daniels, 172 Mich App 374, 378; 431 NW2d 846 (1988). Moreover, this Court has stated that positive identification by witnesses may be sufficient to support a conviction of a crime. See People v Malone, 193 Mich App 366, 371; 483 NW2d 470 (1992), aff’d on other grounds 445 Mich 369; 518 NW2d 418 (1994).
Defendant first argues that reliable identification was impossible because the perpetrator escaped. Defendant cites no authority for this claim, thereby abandoning it on appeal. People v Battle, 161 Mich App 99, 101; 409 NW2d 739 (1987). Defendant further claims that Settles’ identification of defendant to the prosecutor right before the start of trial was “suggestive and invalid” because defendant “was on trial for the offense and there were no other suspects.” Even if defendant had preserved this issue by a motion to suppress Settles’ identification testimony, see People v Lee, 391 Mich 618, 626-627; 218 NW2d 655 (1974), we would not reverse.
Defendant relies on United States v Wade, 388 US 218; 87 S Ct 1926; 18 L Ed 2d 1149 (1967), and People v Anderson, 389 Mich 155; 205 NW2d 461 (1973). The specific issue in those cases was whether a suspect *701was entitled to the presence of counsel at state-compelled, pretrial identification confrontations such as corporeal lineups or show-ups, Wade, supra at 228-230, or at photographic lineups, Anderson, supra at 186-187, situations significantly distinguishable from the case at bar wherein the witnesses identified defendant on the basis of having seen him commit the crime and, in the case of complainant, having additionally seen him in public on three subsequent occasions.
Nonetheless, the Wade and Anderson Courts discuss at great length the unreliability of eyewitness identification and the potential thereby for misidentification:
[T]he annals of criminal law are rife with instances of mistaken identification . . . [t]he identification of strangers is proverbially untrustworthy . . . [t]he influence of improper suggestion upon identifying witnesses probably accounts for more miscarriages of justice than any other single factor ... [suggestion can be created intentionally or unintentionally [and] the dangers for the suspect are particularly grave when the witness’ opportunity for observation was insubstantial. [Wade, supra at 228, 229.]
“[W]e find there are serious problems concerning the accuracy of eyewitness identification and that real prospects for error inhere in the very process of identification completely independent of the subjective accuracy, completeness or good faith of witnesses.” Anderson, supra at 180; see id. at 172-180.
In addition to defendant’s specific claim regarding Settles’ “suggestive” identification before trial, defendant relies on the statements from Wade and Anderson regarding the unreliability of eyewitness testimony to also generally challenge both Settles’ and complain*702ant’s in-court identification of him as the peipetrator. Defendant is not arguing merely that Settles and complainant lack credibility. Rather he is arguing that, when the issue is identification, eyewitness testimony is simply too unreliable to support a conviction notwithstanding the credibility of the witnesses. We utilize the same analysis for both of defendant’s claims.
In People v Gray, 457 Mich 107; 577 NW2d 92 (1998), a rape victim tentatively identified the defendant in a corporeal lineup and then later confirmed her identification when a police officer came to her home and, allegedly to allay her fears, told her a suspect had been arrested and showed her a photograph of the defendant. The Court found that the officer’s action was impermissibly suggestive, but found that there was an independent basis for the victim’s identification of the defendant. The Court first noted that the remedy for an unduly suggestive identification procedure is suppression of the in-court identification unless there is an independent basis for its admission. Id. at 114, n 8; see Wade, supra at 239-240; People v Kurylczyk, 443 Mich 289, 303; 505 NW2d 528 (1993) (Griffin, J.). The Court noted that the independent basis inquiry is a factual one and that the validity of a victim’s in-court identification must be viewed in light of the totality of the circumstances. Gray, supra at 115, citing Neil v Biggers, 409 US 188, 199; 93 S Ct 375; 34 L Ed 2d 401 (1972).
The Court then reiterated the factors, originally set forth in the nonbinding case of People v Kachar, 400 Mich 78; 252 NW2d 807 (1977), that a court should weigh in determining if an independent basis exists for the admission of an in-court identification: (1) prior relationship with or knowledge of the defend*703ant; (2) opportunity to observe the offense, including length of time, lighting, and proximity to the criminal act; (3) length of time between the offense and the disputed identification; (4) accuracy of description compared to the defendant’s actual appearance; (5) previous proper identification or failure to identify the defendant; (6) any prelineup identification lineup of another person as the perpetrator; (7) the nature of the offense and the victim’s age, intelligence, and psychological state; and (8) any idiosyncratic or special features of the defendant. Gray, supra at 116; see also CJI2d 7.8. In Kurylczyk, supra at 307-308, the Court also noted that delays as long as eighteen months after a crime do not necessarily invalidate an eyewitness identification.
In Gray, the nighttime assault lasted approximately an hour, the victim described the five-foot, eight-inch, 180-to-190-pound defendant as a stocky six feet, and her initial identification was tentative. However, the Court found that the improper photographic display was relevant to the weight of the victim’s testimony rather than its admissibility and concluded, primarily on the basis of the victim’s opportunity to observe the offense and her previous proper, albeit tentative, identification of the defendant, that there was a sufficient independent basis for the admission of her in-court identification. Gray, supra at 117; see Wade, supra at 241.
Applying the Gray factors first to Settles’ identification of defendant: Settles had never seen defendant before and initially observed from his vehicle what he thought to be a purse snatching because defendant was standing over a woman who was seated in a car and appeared to be pulling her purse. Settles drove to *704where he could get a better look, got out of his vehicle, and then got a “very good look” at defendant’s face from twelve to twenty feet away while confronting him for “a good minute” before defendant ran away. There was a nine-month delay before trial, and Settles described defendant in his report only as an unknown black male, approximately five-foot, nine inches, or five-foot, ten inches, and 140 to 160 pounds, whereas defendant was actually five-foot, six inches, and weighed 130 pounds at the time of the commission of the crime. While a significant amount of time had passed since the crime and while Settles had a relatively short time to observe defendant, he is a trained police officer whose initial description was accurate and who confronted defendant from twelve to twenty feet away, ordering defendant to freeze. Hence, even if defendant had moved to suppress Settles’ in-court identification and even assuming that Settles’ identification to the prosecutor immediately before trial was impermissibly suggestive, we would conclude that there was a sufficient independent basis for Settles’ identification of defendant. In any event, the trial court found that Settles’ identification of defendant was merely corroborative of complainant’s identification. Further, defense counsel rigorously cross-examined Settles regarding his identification of defendant. See Wade, supra at 232.
Complainant also had never seen defendant before the crime and, even though she testified that it was “fairly [ ]dusk, dark,” she had approximately five minutes to talk to him and observe his face from a foot away in a parking area right next to and lighted by the store she had just left. She initially described defendant to the police as having a medium complex*705ion, five-foot, nine inches, and 170 pounds, a description that was even further from defendant’s actual appearance than Settles’ description. However, as the trial court noted, complainant testified that defendant was wearing a dark, bulky jacket that may have made him seem heavier and, because she was seated in her car and defendant was standing over her, he might have seemed taller. Importantly, she subsequently and coincidentally saw defendant in the area, in public, within a few days of the offense and on two additional occasions, the second of which resulted in his arrest approximately three months after the robbery. She attended no lineups, never identified any person other than defendant as the perpetrator, and testified that she was very frightened during the robbery, but there was no evidence that she was under the influence of any drugs or was lacking intelligence. The trial court mentioned but did not address the issue of defendant’s complexion, the evidence regarding which was indeterminate.
On the basis of complainant’s face-to-face opportunity to observe defendant commit the crime and her subsequent identification of him in public on three additional occasions, one of which led to his arrest, we conclude that there was an independent basis for complainant’s identification of defendant, notwithstanding the unreliability of eyewitness testimony in general. Any discrepancy between complainant’s initial description and defendant’s actual appearance is relevant to the weight of such evidence, not to its admissibility. Gray, supra. Considering the evidence in the light most favorable to the prosecution, the evidence was sufficient to support defendant’s conviction beyond a reasonable doubt. Johnson, supra. Fur*706ther, we decline to read the language in Wade, supra, or Anderson, supra, regarding the unreliability of eyewitness testimony as a general proscription against the use of such testimony.
Affirmed.